Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

contribution to computer aided design evaluation utilizing flexible estimation and multiple linear regression… Forde, Bruce W. R. 1984

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1984_A7 F67.pdf [ 3.76MB ]
Metadata
JSON: 831-1.0062467.json
JSON-LD: 831-1.0062467-ld.json
RDF/XML (Pretty): 831-1.0062467-rdf.xml
RDF/JSON: 831-1.0062467-rdf.json
Turtle: 831-1.0062467-turtle.txt
N-Triples: 831-1.0062467-rdf-ntriples.txt
Original Record: 831-1.0062467-source.json
Full Text
831-1.0062467-fulltext.txt
Citation
831-1.0062467.ris

Full Text

A CONTRIBUTION TO COMPUTER AIDED DESIGN EVALUATION UTILIZING FLEXIBLE ESTIMATION AND MULTIPLE LINEAR REGRESSION ANALYSIS  by  BRUCE W.R.  FORDE  A THESIS IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE  in  THE FACULTY OF GRADUATE STUDIES (The Department o f C i v i l  Engineering)  We accept t h i s t h e s i s as conforming to the required standard  September, 1984  © Bruce W.R.  Forde  In p r e s e n t i n g  this thesis  requirements British  it  freely available  for  f u l f i l m e n t of the  f o r an a d v a n c e d d e g r e e a t t h e U n i v e r s i t y  of  Columbia,  agree t h a t  in partial  I agree that f o r reference  permission  the L i b r a r y  shall  and s t u d y .  I  thesis  s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e h e a d o f my  understood  that  copying or p u b l i c a t i o n  f i n a n c i a l gain  shall  Department  of  CIVIL ENGINEERING  The U n i v e r s i t y o f B r i t i s h 1956 Main Mall V a n c o u v e r , Canada V6T 1Y3  O  c  t  o  b  e  r  12  >  !9  8 4  of t h i s  Iti s thesis  n o t be a l l o w e d w i t h o u t my  permission.  Date  further  f o r extensive copying of t h i s  d e p a r t m e n t o r by h i s o r h e r r e p r e s e n t a t i v e s . for  make  Columbia  written  ABSTRACT Improvements  in  fabrication costs, may  design,  achieved  through  a  better  understanding  of  cause an o v e r a l l cost reduction for s t e e l structures.  A f l e x i b l e estimation routine, able to s a t i s f y the needs of the particular user, has been developed  to provide the mechanism for design evaluation.  The  functions used by the estimation program are provided by a Multiple Linear Regression The  (MLR)  analysis of data collected by an Information System (IS).  integration of f a b r i c a t i o n control and analysis provided by this system  permits  its  implementation  in  existing  environments,  important technological gain for the fabrication process. approaches which have r e l i e d heavily on experience  and  presents  an  T r a d i t i o n a l design  can now  be evaluated and  improved i n terms of cost competitiveness, prior to or during f a b r i c a t i o n , by the proposed MLR strategy.  - ii -  TABLE OF CONTENTS ABSTRACT  i i  TABLE OF CONTENTS  i i i  ACKNOWLEDGEMENTS 1.  iv  INTRODUCTION  2.  1.1  Background Information  1  1.2  Research at U.B.C  2  STEEL STRUCTURES - Design and Fabrication  3.  2.1  Design E f f i c i e n c y  2.2  Fabrication Costs  6  2.3  Erection Costs  9  2.4  Current Practice  10  2.5  Design/Fabrication Interaction  13  ,  5  COMPUTER AIDED DESIGN EVALUATION 3.1  The Design Evaluation Program (ESTImate)  3.2  Multiple Linear Regression and the Information System 3.2.1 3.2.2 3.2.3 3.2.4  3.3  Motivation for MLR The Information System Example Use of MLR MLR - Design and Theory  17 21 24 26 30  Applications of Program ESTImate 3.3.1 3.3.2  , 3.3.3 4.  Page  Design Optimization Improvement of the Fabrication and Estimation Processes Research and Education  34 36 37  SUMMARY 4.1  Conclusions  38  4.2  Extensions of this Research  39  REFERENCES  40  APPENDIX  42  - iii -  ACKNOWLEDGEMENTS The  investigations and research  possible by a Natural  Sciences  contained  and Engineering  i n this  thesis were made  Research Council of Canada  (NSERC) postgraduate scholarship. Manfred Frank, Chief Engineer INC.,  and Siegfried  Civil  Engineering  F. Stiemer, at  of the Western Bridge Division of CANRON Associate Professor  the University  of B r i t i s h  i n the Department of Columbia,  who  was  my  supervisor, are due sincere thanks for their guidance and cooperation. I am e s p e c i a l l y grateful to Elaine Lee, my friend, who provided me with motivation throughout my research.  - iv -  l.  1. INTRODUCTION 1.1  Background Information Over the  steel  past  few  i n Canada and  the United  based on the experience independent studies. material  weight  by  decades, the design and States has  fabrication  which  was  the results of a  few  Improvements i n design were found Hoerner  (1982)  structural  followed a procedure  of the design profession and  von  of  [18], and  by minimization of  Schmit  (1960)  [16],  by  r e p e t i t i o n of similar shapes by Moses and Goble (1970) [11], and by stimulation [7].  of design/fabrication Interaction by Forde, Until  now,  most  of  these  efforts  Leung and  have f a l l e n  Stiemer  short  (1984)  of providing  "minimum cost designs", simply because the costs incurred during fabrication have been only approximately  known.  The  cause of this uncertainty can  be  attributed to the nature of the design and fabrication processes which make each project unique. The problem facing fabricators i s to i d e n t i f y where costs are incurred, to  predict  these  process.  costs  Recording  for future  fabrication  jobs,  times  and  and  to  control  process  necessary  for the estimation process, has been found  expensive  to  there  has  justify  been  a  the  information that  trend  towards  costs,  monitor  In  the  the  detail  to be too tedious and  i t provides.  collecting  and  For  information  this on  a  reason, global  perspective. Recent economical  developments data  in  storage and  microcomputer analysis.  The  technology  offer  addressable  memory and  c a p a b i l i t y of many systems e a s i l y s a t i s f y the requirements fabricators.  The  simplifications,  and  improved  a  means  of  storage  of designers and  efficiency,  that  are  2. p o t e n t i a l l y available from a computerized f a b r i c a t i o n process, have provided the predominant motivation for research i n this area.  1.2  Research at U.B.C. An  extensive  research  project has  been started  at  the  University  of  B r i t i s h Columbia i n the areas of design evaluation and f a b r i c a t i o n analysis. One  goal  of this  project i s to e s t a b l i s h a mechanism which integrates  the  design and  f a b r i c a t i o n processes so that a better understanding of costs i s  obtained.  To  meet this  goal,  several smaller projects must be  completed.  Fabrication analysis can be divided into the c o l l e c t i o n , analysis, and back of f a b r i c a t i o n data. flexible  estimation  Design evaluation requires  program,  structural  analysis  modules, and improvement or optimization routines.  feed-  the development of a and  code  requirement  Further projects that are  required i n order to assure integration of the whole include several modules that provide user f r i e n d l y Interface for Interactive input, and compatibility for transfer of information between modules. This  thesis provides  research i n two  an overview of the  t o t a l project, with  extensive  areas:  (1)  Estimation of f a b r i c a t i o n times;  (2)  Analysis of f a b r i c a t i o n data.  Design evaluation In terms of cost can be done using the estimation program which i s described breaks involved  projects  i n this . thesis (ESTImate).  down into  basic  elements,  i n performing a set of standard  This program systematically  estimating  the  time  f a b r i c a t i o n operations.  and  cost  Owing to  the i n d i v i d u a l i s t i c nature of each f a b r i c a t i o n plant, the estimation programs used  by  various  contractors  are  not  the  same.  Since  there  exists  no  universal standard, nor should there as this would exclude the p o t e n t i a l for improvement, a f l e x i b l e has  estimation program i s e s s e n t i a l .  been developed by adopting  a fourth generation  Program ESTImate  approach to programming  [13] which makes i t exceptionally v e r s a t i l e for the construction of estimation equations. Factors contributing to the requirement of f l e x i b l e estimation are: (1)  The  fabrication  process  i s not  a mass production,  so  the  estimation  routine must be able to accept a wide variety of projects. (2)  Fabrication plants  are  application oriented, with  some being  better  equipped to do s p e c i f i c projects than others, hence one must allow for a variety of f a b r i c a t i o n (3)  techniques.  Design evaluation must be done i n a detailed manner, as provided by an estimation program, so that improvement or optimization can be The  conversion  of information which i s gathered  from a global  achieved. perspec-  t i v e , to detailed information which i s required for estimation purposes, can be done through the use of a comprehensive information system by Berry (1984) [2] coupled with a multiple linear regression program. The implications a r i s i n g from the use of MLR (1)  Information  are:  need only be collected at "checkstations" which are located  at key locations throughout the plant. (2)  Integration of scheduling, c o n t r o l l i n g , and estimating i s made possible through the use of an information system which c o l l e c t s this data at the global l e v e l .  (3)  The  fabrication  plant  can  be  evaluated  i n terms  of  performance  and  economy by analyzing the data collected by this information system over a period of time.  4.  The multiple l i n e a r regression module that has been developed as part of this thesis i s capable of analyzing data collected from a global perspective only.  Establishment  of intermediate checkstations w i l l provide information  which requires a more sophisticated analysis.  2. STEEL STRUCTURES Design 2.1  and  Fabrication  Design E f f i c i e n c y The  e f f i c i e n c y of a s t r u c t u r a l design can be defined as the extent to  which i t s a t i s f i e s s t r u c t u r a l , a r c h i t e c t u r a l , and cost constraints. The role of the structural engineer  i n the design process has t r a d i t i o n a l l y been to  choose and design the most e f f i c i e n t structural system subject to the given constraints.  This has been no easy task, since the constraints are linked so  that there must be some tradeoff made i n reaching the optimal design. tectural requirements  Archi-  are generally defined by the owner as a given require-  ment of the project, so s t r u c t u r a l form i s to be varied to provide the most economical design.  Another important fact complicates the s i t u a t i o n : i n most  cases, the designer w i l l not have access to the necessary cost data; the  formulation  function,  and  of  the  the  design  designer  objective can  i s unable  to  not  perform  be  i n terms  formal  of  hence, a cost  structural  cost  optimization. In the past, "experience" has been the key to e f f i c i e n t s t r u c t u r a l s t e e l design. advantage  Successful engineers have established design guidelines which take of  fabricators.  Information  that has  However, i n an  been acquired  effort  to maintain  through  Interaction with  confidentiality  of  their  direct costs, fabricators are l i k e l y to provide the designer with information of only a q u a l i t a t i v e nature.  The fabricators have a more quantitative look  at the information since they are able to d i r e c t l y view the operation and can account for the costs as they are incurred.  Nevertheless, sometimes even the  fabricators are unsure of the exact costs associated with various structural forms, due to the v a r i a t i o n of the type and quantity of work done for any project.  one  6.  The  size and complexity  structural inherent the  steel  of the projects that  fabricators vary  significantly  are undertaken by most  over  time.  I t i s this  v a r i a b i l i t y of the s t r u c t u r a l s t e e l f a b r i c a t i o n process which poses  largest problem  to the designer.  In order  to assign  a cost  to an  a r b i t r a r y s t r u c t u r a l form, costs f o r each component can be calculated from a standard  set of f a b r i c a t i o n variables, which  are i n turn  determined and  updated through analysis of past projects.  The f a b r i c a t i o n process, although  composed of p h y s i c a l l y simple operations,  as a whole i s extremely complica-  ted;  hence, associating costs  with the f a b r i c a t i o n of s p e c i f i c s t r u c t u r a l  components  is difficult.  In fact  the volume  collection,  organization,  and analysis  of work  of data required  Involved  to i d e n t i f y where  costs are incurred has e f f e c t i v e l y eliminated this approach from practice.  i n the  conventional  To surpass this b a r r i e r , the designer must investigate the nature  of the costs associated with the f a b r i c a t i o n process.  2.2  Fabrication Costs The  cost of f a b r i c a t i n g s t e e l structures may be broken down Into  four  major categories: 1.  Material ;  2.  Labour  3.  Equipment;  4.  Overhead .  ;  Steel i s purchased from a m i l l additional  charges  also  based  at a unit cost  per weight, plus some  on weight; hence, the cost  approximately proportional to the weight of the structure. major d r i v i n g force f o r minimum weight design,  of material i s  This has been the  since minimizing the weight  7.  also  minimizes  the material cost.  Minimum  weight  approaches,  although  academically appearing to be reasonable, have been found to provide designs which are often impractical [10].  1  UoH  Figure 1 .  I sa'.J  "minimum  "tight'  Minimum weight •*• Minimum cost?  Labour and equipment costs are not so e a s i l y related to the form of the structure.  Obviously some designs are more e f f i c i e n t than others i n terms of  ease of f a b r i c a t i o n , but i t i s d i f f i c u l t to determine why.  A key factor i n  these costs i s material handling, which appears to be one area i n the f a b r i cation process which has large potential  f o r improvement.  Handling costs  are presently assumed to be some percentage of the costs d i r e c t l y related to the f a b r i c a t i o n operation.  The reduction i n costs associated with r e p e t i t i v e  s t r u c t u r a l forms may be attributable to improvement i n productivity for many operations  i n the f a b r i c a t i o n  time, and labour e f f i c i e n c y .  process,  such  as material handling, set-up  Designers t r y to account f o r this by using as  8.  many s i m i l a r components as possible i n a structure.  The  r i s i n g labour  machinery costs are becoming an increasingly important part of the  and  overall  cost, so a more deterministic approach to their evaluation i s needed. The  f a b r i c a t i o n cost of the i t h s t r u c t u r a l member type can be written  as: i - N [E c W j=l j n  c 1  where  c ,c„ m a  m  3  °i + E c T ] k-1 k  (1)  k  are unit costs for material and  labour,  i s the weight of material j .  This  T^  i s the time required to perform operation k.  n^,o^  are the number of components and operations for member i .  N  i s the number of i d e n t i c a l members.  equation  first  appears to be  component represents  the standard  i n use  the material cost  by most f a b r i c a t o r s .  The  ( i n terms of weight) and  the  second comes from the direct labour costs ( i n terms of time). Typical sawing,  operations  burning,  used  punching,  In  the  drilling,  f a b r i c a t i o n process making  cleaning, painting, handling, machining, etc. estimate Leung  the times associated with  [ 9 ] , and  appendix. the  The  are  used as  templates,  for the  hence,  an  Interaction  welding,  have been developed  example problem shown i n  quantity of material fabricated has  productivity;  fitting,  shearing,  Equations which can be used to  these operations  a basis  are:  diagram  by the  an unknown influence on must  be  developed  from  experience  r e l a t i n g N to T.  operations  of the f a b r i c a t i o n plant so that a time-history plot of the work  being done i s a v a i l a b l e .  One method of doing this i s to monitor the d a i l y  This information can be used to i d e n t i f y the rate  9. at  which  work i s proceeding  functions  of time,  which  on various  components  can be translated  into  of the structure as  curves  which  represent  productivity improvement due to r e p e t i t i o n of similar acts. The t o t a l cost of a structure can be written as the sum of the component costs: Total Cost =  E(C.) + 0 J  (2)  C^ =  cost of component i (as given i n equation 1)  0  overhead costs  =  Examining these equations reveals that: c^ and c^, are known constants; W and T are known functions of N; and N i s the variable that we can manipulate to reduce the costs.  2.3  Erection Costs The  current  procedure  for estimation  uncertainty associated with i t s accuracy.  of erection  costs  has a large  The estimated cost of erecting an  a r b i t r a r y structure has commonly been based on the material weight, without provision  f o r the complexity  structures  of the erection procedure.  are more erection f r i e n d l y than others;  w i l l a l t e r their costs by a factor from experience.  Obviously,  some  hence, some fabricators Decisions of this nature  have been i n error by as much as 100% and more due to the large variations i n productivity associated with the erection a c t i v i t i e s  [9].  These variations  may be attributed to the same kind of improvement trends which result from r e p e t i t i v e a c t i v i t i e s as encountered i n the f a b r i c a t i o n process. The outline  proposed procedure f o r evaluating erection costs follows a similar as f o r f a b r i c a t i o n costs.  Costs  associated  with  an a r b i t r a r y  10.  structure w i l l during  be  calculated from a  set  of  standard  activities  performed  erection (transportation, sorting, l i f t i n g , b o l t i n g , e t c . ) .  A large  portion of equipment used during erection i s rented; hence, equipment cost i s proportional to the erection time. tion  process where most equipment  e s s e n t i a l l y considered  the  This Is quite d i f f e r e n t from the f a b r i c a i s owned by  the  can  be  as overhead.  The programs developed i n this thesis are primarily intended  for use i n  f a b r i c a t i o n process;  costs  however,  they  are  flexible  erection and other construction a c t i v i t i e s can also be  2.4  fabricator and  so  that  for  incorporated.  Current Practice Most structures  that  are  follow a process which has economical reasons. are:  the  owner, a  consultant  to  The  constructed  the United  States  been established for h i s t o r i c a l , p o l i t i c a l ,  and  three parties involved i n the construction process  consultant,  design  i n Canada and  the  and  a  contractor.  structure, and  to  performs the construction of the structure.  The  supervise  owner employs the  contractor  the who  Since the design i s p a r t i a l l y or  f u l l y complete at the time when the contract i s put up for tender, the  design  usually w i l l have been done without any i n t e r a c t i o n between the designer  and  the contractor.  can  foresee  the  This kind of practice i s only acceptable  costs  associated  with  various  design with minimum cost can be chosen. of  the cost i s incurred  only a l i t t l e knowledge. unable  to  design  the  i f the designer  s t r u c t u r a l forms,  so  For s t e e l structures, the  that  the  majority  i n a f a b r i c a t i o n plant, of which the designer  has  In this case i t i s apparent that the consultant i s  most  cost  interaction with the f a b r i c a t o r .  efficient  building, unless  there  Is some  11.  The i n t e r n a l organization of most s t r u c t u r a l s t e e l f a b r i c a t i o n companies follow the functions shown i n Figure 2 as adapted from a paper by Adlard [1]. Three  categories  engineering  are  of  interest  a l l have equally  to  important  us.  Business,  roles i n the  production,  o v e r a l l construction  process, yet there has been a trend i n the past towards concentrating in  areas where productivity i s quantifiable.  method for evaluating against  Since  the business aspects of one  the rest of the corporation's  policy are usually formulated  and  there exists a operation,  by  effort standard  comparison  p o r t f o l i o , decisions regarding company  at this l e v e l .  STRUCTURAL STEEL FABRICATION BUS I K I S S  Resource Manages* a t  1) 2)  PUnt Layout Inventory  3)  Scheduling  4)  Manpower Utilization  5)  Strategic Planning  KMCINIERIKG  PRODUCTION  Product ion Control  Economic Accounting  t)  Coat Control  2)  Cost Est i n a t i o n  t)  4) G e n e r a l Accounting  1)  Inventory  2) R o u t i n g  3) P r o c e s s Analysis  Document Preparation  3) M a t e r i a l  /  Scheduling Handling  2)  Drafting  3)  Material Bills  4)  Data Management  4 ) QC/QA 5)  Maintenance  Parts Database  1)  Production Cost Data Acguisi tion  2)  Production Product i v ity Data Acquisition  6) P u r c h a s e / R e c e i v e 7) D a t a  Analysii  Testing  1) D e s i g n Standards 2) D e s i g n Bvaluat ion 3) P r o d u c t i o n Bvaluat ion 4)  Replacement 1nvestment Analysis  Management  * QC/QA - Quality Control, Quality Assurance.  Figure 2.  Organization  of Structural Steel Fabrication Functions.  S t r i c t short terra cost j u s t i f i c a t i o n methods based on return on investment have been shown to provide  inadeqate analysis of decisions  process expansion and improvement [1]. wrong method  of  analysis  i s used,  regarding  The real problem here i s not that the since  surely  a l l good  company p o l i c y  12.  criteria  are linked  i n some way to p r o f i t  maximization,  analysis i s being done without the proper information. involved  i n fabrication  methods of documentation  are linked lead  rather that the  The various functions  as shown i n Figure  3.  Conventional  us to a maze of information contained i n :  progress statements, manpower u t i l i z a t i o n summaries, cost control data, e t c . The inevitable result i s a process which involves much doubling up of e f f o r t and a general lack of communication between the various departments.  Many of  these problems can be eliminated through the use of common databases. would  be possible i n a computer integrated manufacturing  This  environment [1];  however, the scope of this thesis i s limited, so the implementation  of this  process must be l e f t for others.  FABRICATOR ERECTOR  CONSULTANT DESIGNER (1) DESIGN TEAM  Typical Interaction: (1) (2) (3) (4)  Project Documents, Scheduling, Design Revisions/Proposals. Project Documents, Estimates, Strategic Planning, Design Evaluation. Process Data, Shop Drawings, Material B i l l s . Progress Reports, Manpower U t i l i z a t i o n , Scheduling, Inventory. Figure 3.  Current Links to Design and Fabrication Functions.  13.  Assuming  that  the  fabricator  mentioned above, attention should the  one between  structural  steel  the designer  has  the  and  the f a b r i c a t o r .  problems  design  drawings  Presently, from  link:  several  the consultant  A detailed f a b r i c a t i o n cost estimate i s made  from these drawings and submitted back to the consultant bid.  internal  be redirected to another important  fabricators receive  withthe i n v i t a t i o n to tender.  resolved  i n the form of a  At this time the fabricator has i d e n t i f i e d some areas where savings can  be made by suggesting  some s t r u c t u r a l l y equivalent, but more cost  designs f o r c e r t a i n components of the structure.  Quite  efficient  often the designer  may receive notice from the fabricator at the time of the b i d that there may be  some means of cutting costs, since this i s an e f f e c t i v e means of making  his bid more a t t r a c t i v e .  But a f t e r this time, these changes are usually not  suggested by the fabricator due to the i n a b i l i t y of the design  process to  easily accept changes. The  c h a r a c t e r i s t i c s of each f a b r i c a t i o n plant depends on the s i z e , and  type of projects that the fabricator i s involved with.  This means that every  f a b r i c a t i o n plant w i l l have I t s own set of cost variables which make i t more suitable f o r the f a b r i c a t i o n of c e r t a i n types of structures than for others. Thus, interaction between the designer  and the fabricator i s essential f o r  a l l projects, and must become an i n t e g r a l part of the construction  2.5  Design/Fabrication The  proposed  Interaction  implementation  of a  opposed to that currently i n existence the  current  process.  system there  new  design/fabrication  system as  i s shown i n Figures 4a and 4b. Under  i s no r e a l i s t i c  method  f o r design  optimization,  since there i s b a s i c a l l y no d i r e c t interaction between the designer  and the  14. fabricator.  The  u n i d i r e c t i o n a l flow of information i n the construction of  most projects i s marked by  the  inability  of the design process  to accept  changes (whether b e n e f i c i a l or not) as suggested by the f a b r i c a t o r . Experience' gained often prevented  by  the fabricator  during the course of h i s work i s  by the design process from being put to i t s best use.  designer i s unable  The  to accommodate the fabricator's desires once the job i s  awarded because of an i n f l e x i b l e design process, yet the fabricator i s not ready to " t e l l a l l " before this time since this may  give away some competi-  tive advantage. This dilemma cannot be completely overcome i n conventional p r a c t i c e , but a  compromise can  be  made.  Mutually  acceptable  revision  procedures  could  accelerate the flow of information between the two parties so that improved designs  can  cooperation  be  put  forward  involving  at  sharing  the  moment they  of  savings,  and  are  envisioned.  bonuses  for  Further  innovative  advances, could lead to established optimization procedures. The proposed design/fabrication process (Figure 4b) provides a mechanism for f a b r i c a t i o n analysis, which offers continuously improved estimation and design procedures. found  i n the  research  This may  current  topic.  The  process use  of  remove the heavy dependence upon (Figure 4a), such  a  system  but  i t arouses  i n practice  experience  another  large  requires further  investigation and analysis i n terms of the effects that this may have on the contractual obligations of the parties involved i n the construction process. Direct  application  possible  of  the  research contained  for design-build contracts, and  i n this  thesis  i s currently  once the l e g a l aspects mentioned  above are resolved, application w i l l also be made i n conventional practice.  Preliminary Design  EXPERIENCE consultant fabricator  ;ure 4 ( a ) .  Current Design/Fabrication Proce  Figure 4(b).  Proposed Design/Fabrication Process.  COMPUTER AIDED DESIGN EVALUATION  3 .  3.1  The D e s i g n E v a l u a t i o n "Flexibility"  was  c a l l e d ESTImate. provide trary  advances  in  arbitrary the  Program  the  primary  rest  process  compiler  of  referred  and  set  of the  FORTRAN e q u a t i o n s , program,  all  program which  kinds  geared  towards  screen  commands i n t o by  of  the  individually  fabrication a mechanism  and then  from w i t h i n  an  estimation  industry. for  usage.  the  Most  of  Interactive  user,  whether to  their  terms of  for  storage  calculation  languages  entire  syntax  and  (FORTRAN, P a s c a l , in this  an assembly  these  user,  in  expressions, in during  the  firms  generating  program.  This  an  technique and i s  which  can  process  is one  large  and  batch  are  mode  mathematical  the o r d e r of must  be  control  C, e t c . )  where the  have developed programs  programs (VISICALC, CALC,  operations.  a r e a has p a r t i a l l y language,  Recent  e s t i m a t i o n e q u a t i o n s , c o u l d be approached  mathematical  converted  arbi-  c o m p i l i n g and l i n k i n g them to  routine,  computer software  the  the  I  thesis.  of  Several  similar  in  the  provided  of data u s i n g a r b i t r a r y  many ways.  software  found i n  construction  construction  quantities  that  a  to as a f o u r t h g e n e r a t i o n approach to programming [ 1 3 ] ,  The  input  for  and too much coding to accommodate the  data  of the key elements of t h i s  in  design c r i t e r i o n  The use o f c o n v e n t i o n a l programming t e c h n i q u e s was found to  too many r e s t r i c t i o n s  project  17.  based The  or  by  on  the  equivalent.  etc.)  are  translation  individual direct  for  equations  of as  means,  must  be  Both the  logic  of  the o p e r a t i o n s , and the mechanism  internally already  controlled.  available  in  This higher  level  are d u p l i c a t e d by these programs. overcome t h i s  means  Some  redundancy by programming  u s e r must c o n t r o l a l l  functions  of  data  18.  storage and manipulation. its  intended  This approach has been shown to be successful for  i n t e r a c t i v e application; however, i f the program i s to  large quantities of data using the same set of equations,  process  then a great deal  of time and storage w i l l be wasted by redundant operations. The solution to this d i f f i c u l t y i n programming comes from assessing the needs of the user. and  The program w i l l be used on a regular basis by designers  fabricators for cost estimation, and w i l l only be updated when i t has  been shown to be  i n error.  This w i l l  l i k e l y be done at regular intervals  (weeks, months, years) depending on the quantity and type of work being done by  the user.  From this description of user needs, i t i s apparent that the  previously discussed techniques  are much more f l e x i b l e than required for our  purpose, and that another approach i s needed. updated by the user, Into the code.  then one  could simply  This would be acceptable  organization of the program and was rest of the coding. program w i l l associated developed  with  this  to allow  transition.  the  f a m i l i a r with  avoid  be  some loss of  this  a  subroutine  the  depicted  i n Figure 5, does the same kind of interpretation and  the  results  interactive  by  problem,  contains  that  means of a program.  calculator programs do,  i t assembles  a  program  the  the personnel involved with this  there w i l l To  equations)  right  able to properly l i n k his work to the  construction of an estimate  estimation  never to be  incorporate the equations  i f the user was  In most circumstances,  change over time and  If the program was  which  will  but be  rather used  for  familiarity system file  was  (which  This routine, translation  than compute that  the  operation.  "READACT" i s a subroutine which reads estimation functions from the a c t i v i t y database, while "VLE" variable  database.  reads a l i s t of variables and their descriptions from a Subroutine  "CREATE" assembles the estimation functions  19. Into equivalent FORTRAN expressions, checking The  result  the logic of the input data.  i s an estimate subroutine f i l e which contains a complete FORTRAN  subroutine which w i l l be used f o r estimation.  A C T I V I T Y DATABASE - E s t 1mat i o n functions  VARIABLE DATABASE - Variable list - descr i p t ions  VLE  REAOACT  1 i SUBRQUTINE CREATE  ESTIMATE SUBROUTINE FILE  Figure 5. Construction of the Estimate Subroutine  File.  The estimate subroutine f i l e must be compiled and linked  to the rest of  the program to complete the program construction. The preliminary version of ESTImate uses  a VAX  11  run-time  library  function (invoked  by  subroutine  "GETSUB") to do this from within the program; however, l a t e r versions running on UNIX could do this by other means. then  done as  i n Figure  6.  The use of the estimation routine i s  A quantity database,  containing a l l physical  properties and a c t i v i t y methods i s read by subroutine "READQUAN", while is  used to read a control f i l e which defines the content and  output data.  format  "VLE" of the  Subroutine "ESTIMATE" computes a l l of the estimated times/costs  for each a c t i v i t y and writes this information to the estimate  database.  20.  QUANTITY DATABASE Physical properties A c t i v i t y methods  EST I MA T E SUBROUTINE FILE  V A R I A B L E OUTPUT CONTROL output variables  VLE  (  GETSUB  .  I 11  REAOOUAN  SUBROUTINE ESTIMATE  -  Figure 6.  ESTIMATE Estimated for each  DATABASE Times/Costs activity  Using the Estimation Subroutine.  Information contained In the estimate database of  manners.  Consultants may  utilize  can be used i n a variety  this information as a means of design  evaluation i n an optimization routine.  Fabricators can use i t to summarize  t h e i r bid proposals, to analyze t h e i r fabrication plant, or to assess the accuracy of their estimation functions. has  a multiple l i n e a r  estimate and "READIS", and the a c t i v i t y Input  The preliminary version of ESTImate  regression routine (Figure 7, which reads  from  the  information system databases, using subroutines "READESTI" and computes regression c o e f f i c i e n t s which can be used  to update  database.  mechanisms  (QINPUT, ISINPUT) are  not  included i n the  existing  program, but are shown i n Figure 8 to demonstrate their potential use.  The  assembly of the quantity database  can be automated using a tablet or menu  [9],  depending on the preference of the user.  approach,  as shown by Leung  ESTIMATE Estimated for each  INFORMAT I ON S Y S T E M DATABASE - Actual Tiroes/Costs  DATABASE Times/Costs activity  c  READESTI i  REAOIS  i L SUBROUTINE M L R  M L R coefficients - used to update A c t i v i t y Database  Figure 7.  The Multiple Linear Regression Subroutine.  Input to the information system database can be done by cross-referencing the estimate database and  schedule  and  output  the actual times/costs experienced i s dependent upon both  the user's preference and  hardware to be used, so they are not i n s t a l l e d . explained i n d e t a i l by the user's guide i n the  3.2 3.2.1  i n the shop.  Bid the  Use of program ESTImate i s  appendix.  Multiple Linear Regression and the Information System Motivation for MLR With a primary  concern  estimation parameters,  of assessing the accuracy  of i n d i v i d u a l  cost  s o l e l y from information which has been collected on a  global basis, a systematic method of analysis i s needed.  This system must be  (D T3 O 1 rt HO 0  W  c  r| (D CO  •  ,3  CT rt =r  O to CO H» 0°  CO  W M  H> CO  rt  o  a i-«  o OQ 1 P> 5  •  VARIABLE INPUT CONTROL - Input variables - Input descriptions  \  i  ESTIMATE SUBROUTINE FILE kGETSUB f  (D 3 O M O CO (I> CT v: rt  a  4  a>  QUANTITY INPUT CONROL - input format  CI5C i  SUBROUTINE OINPUT  \  VLE  1  ( _  SUBROUTINE CREATE  VARIABLE OUTPUT CONTROL -output variables  CISC DATABASE -Sect ion propertles  i  VLE  \ * — — INTERACTIVVE I INPUT  1  • REFERENCE to Estimate  ESTIMATE DATABASE Estimated Times/Costs for each activity  (by Est(mat  INFORMATION SYSTEM INPUT CONTROL - input format  READQUAN  SUBROUTINE ESTIMATE  or)  J  QUANTITY DATABASE - Physical properties - Activity methods  rt n> CO H01  cp>  VARIABLE DATABASE - Var(able 1 1st - descriptions  1  o o 9 •a M (B rt (t> Cu  Br  <  ACTIVITY DATABASE - Est(mat ion funct tons  f  ( \  INFORMATION SYSTEM DATABASE - Actual Times/Costs  SUBROUTINE ISINPUT  INTERACTIVE INPUT (by Shop)  A* J  \  Not  j Completed  \ READFSTI  READIS  3  SUBROUTINE MLR  /  SUBROUTINE SCHEDULE  o rt rt  n  M  MLR coeffIcients - used to update Activity Database  SCHEDULE OUTPUT Progress reports Schedu1es  s(0 rO  23.  able to break down times (or other c h a r a c t e r i s t i c parameters) measured at the entrance  and exit  from a shop to determine how much e f f o r t  various  operations  during  regression i s a technique  the  fabrication  process.  The  linear  DETAILED INFORMATION Estimation functions.  The Conversion  from Global to Detailed  Information.  mechanism f o r c o l l e c t i n g this global data, and the best locations  for establishment by  Multiple  that can perform this kind of analysis.  GLOBAL INFORMATION Times/Costs measured i n the shop.  Figure 9.  was spent on  analyzing  of "checkstations" i n a p a r t i c u l a r plant, can be determined  the combination  fabricator  i n terms  collection  of data  of  of existing  scheduling,  f o r MLR  processes  controlling,  i s intended  and the needs of the and  to take  estimating.  advantage  of  The current  procedures and control methods rather than to impose a new system i n place of the  old.  The existing  control system may c o l l e c t  a l l of the information  required by MLR, or i t may need some discrete modifications to upgrade i t to that l e v e l .  The i n d i v i d u a l i s t i c nature  of fabrication plants, complimented  with the need to monitor and control the fabrication process, accentuates the u t i l i t y of multiple linear regression analysis. The arises  concept of an information system based on a checkstation approach  f o r psychological and economical reasons.  Monitoring  of individual  24. a c t i v i t i e s i n the shop i s unwanted by the workers, since this would mean that the productivity of each worker could be investigated, nor management  since  the  cost  and  operations would be p r o h i b i t i v e .  the  time associated  The  i s i t wanted by  with keeping  track  of  ideal s i t u a t i o n offered by this system  would be to e s t a b l i s h a few key locations i n the f a b r i c a t i o n plant from which global  data  pertaining  to  item  gathered at these checkstations which would scheduling  evaluate and  the  progress can  accuracy  c o n t r o l l i n g data  would  be  extracted.  The  then be converted to o f f e r of  for  the plant  estimation  process,  operations,  and  to  data  information to  provide  assess  the  This approach to information c o l l e c t i o n , organization, and analysis  can  fabrication plant i n terms of i t s productivity and economy.  provide  both the short-term  to meet, as  well  as  the  improvements that the fabricators are long-term technological competitiveness  obligated that  the  North American industry i s lacking [8].  3.2.2  The Information System An  information  system i s a planning  on the c o l l e c t i o n , organization, and  SHOP Performance of the fabrication operations  Figure 10.  and  control process which i s based  analysis of data u t i l i z i n g a computer.  GLOBAL INFORMATION Times/Costs measured i n the shop.  The Information System (IS).  25. This system, as described by Berry [2], w i l l gather information from the shop on a regular basis so that i t may The programs developed an  arbitrary  WBS  so  that  be used to analyse the f a b r i c a t i o n process.  i n this research project are able to accommodate they  can  be  applied  to  existing  fabrication  operations. The  implementation  of  an  information  system  in  a  non-computerized  environment i s l i k e l y to be done over some period of time; hence, i t must be compatible  with  maintained  during the t r a n s i t i o n period.  to  have the  the present manual system so that regular operation can be  ability  to  accept  and  The requirement  transmit  of compatibility i s  information  that  is  normally  handled by the present system without requiring extensive changes at the base level.  This can be achieved by doing the replacement i n a modular nature.  Care must computerized cies  be  taken  to  avoid  simply  form, since this may  inherent  in  the  manual  duplicating the  accentuate  system  while  present  system i n a  the l i m i t a t i o n s and ignoring  the  inefficien-  newly available  computer graphic aids (tablet and menu I/O mechanisms). The  operation and  e f f i c i e n c y of an information system i s dependent  the "language" of the data communication. represents  the  fabrication  process  A work breakdown structure, which  i n code  form, must  be  implemented  f a c i l i t a t e e f f e c t i v e communication between users of the system. of  a  work breakdown  structure must  be  independent  of  the  The  translated  into  l e v e l s of production.  code should  Information  be meaningful to the users,  to  design  organization-  r e s p o n s i b i l i t y structure, so that information pertaining to one item has same meaning at various  on  which has  so that  the been  the work  breakdown structure can be associated with physical operations i n the f a b r i c ation process.  26. The data collected using the work breakdown structure can be organized to provide two types of information: project data, and process data.  Project  data pertains to a p a r t i c u l a r project or structure, and w i l l be organized In a structure database. structure  database  Each item (consisting of N i d e n t i c a l pieces) i n the  will  be  tracked  through  progress at key operation checkpoints.  the  plant by  monitoring  item  A time/progress record for each item  w i l l be available to the f a b r i c a t o r , so that a deterministic analysis of the costs incurred during the f a b r i c a t i o n of various structural components may  be  performed. Process  data  pertains  to  a l l plant  processes  drilling/punching, welding, material handling, e t c . ) . from  the  same monitoring  operation as  project  (cleaning,  cutting,  This data i s assembled  data, but  Is organized  to  provide information about the e f f i c i e n c y of the plant i t s e l f .  Simulation of  the  evaluate  fabrication  plant  operation  is  required  in  order  to  the  e f f i c i e n c y of the plant configuration, and with the use of the process data examination of expansion or replacement  investments are made possible.  Thus,  an i n t e r a c t i v e plant optimization can be done by Investigating each a c t i v i t y i n the f a b r i c a t i o n process. level  provides  purchasing, could  then  information which would  shop, be  Monitoring the job progress from the component  used  shipping, and to  be  managerial  interactively  used  by  engineering,  operations.  optimize  the  drafting,  This information  processing of material  through the plant.  3.2.3  Example Use of MLR The  use  of a multiple linear  regression model could be applied i n a  variety of manners; however, i n the following case study a "material b i l l  27. approach" was adopted.  The actual data used for t h i s example i s presented i n  the ESTImate User's Guide on pages 12-24 material b i l l investigated collected shop.  was  (see Appendix).  considered independently.  (shear,  saw,  only at the  burn,  entrance  drill,  and  This choice of example was  exit  Each item on a  Five d i f f e r e n t operations were  punch),  and  the  information  from this p a r t i c u l a r  was  part of  the  purely for the sake of s i m p l i c i t y ,  and  would l i k e l y have to be expanded for commercial use. The  schematic  drawing  Figure 11 represents  of  a  the process  given pre-designated  portion of a f a b r i c a t i o n under consideration.  plant  shown i n  Incoming items  paths which they w i l l follow through t h i s area.  are  Again  some simplifications have been made to c l a r i f y the example.  Figure 11.  Before bill times  will  the MLR  subroutine  analysis can be  be used by  to perform  the  receives  Example Shop Area.  done, data contained  the estimation subroutine fabrication an  input  operations.  file  from  the  In t h i s material  to calculate the Then the  estimation  estimated  information system subroutine,  which  28. contains the estimated times to perform the operations required for each item on the material b i l l .  Through Interactive input from the user i t compiles a  d a t a f i l e which contains the actual times measured In the shop. Input to the MLR  program includes the estimated  times for each item to  complete a set of f a b r i c a t i o n operations ([X]), and the times measured by the information system that were a c t u a l l y experienced the MLR  i n the shop (Y).  Within  program the regression c o e f f i c i e n t s (b) and the r e s i d u a l differences  (U) w i l l be determined.  Choosing this formulation provides output which i s  e a s i l y understood, since a regression c o e f f i c i e n t of b^= 1.00 the actual time experienced as that which was  indicates that  i n performance of operation i Is exactly the same  estimated.  This provides information which allows for easy  adjustment of the estimation parameters, since the corrected value i s simply obtained by multiplying the regression c o e f f i c i e n t by the old value. To test the MLR  program the IS data was a r b i t r a r i l y adjusted so that the  actual times required to perform a c t i v i t i e s 1, 3, and and punching) to  were a l l increased by 10%.  5 (shearing,  Variations of this sort may  burning, appear  be semi-random to the untrained eye, yet they should be e a s i l y recognized  by the MLR values of b  program. 1  = b  3  = b  This i s indicated by the MLR 5  = 1.10,  and b  2  = b^ =  program since i t returned  1.00.  The steps used i n the following example were: (1)  Input data  Is summarized  i n a material b i l l  ( a l l physical properties  used by the estimation program). (2)  Estimated  times [X] for each a c t i v i t y are calculated by the  estimation  program for each item on the material b i l l . (3)  Actual global times (Y) are measured by the information system for each item on the material b i l l .  ESTIMATION FUNCTIONS  29.  Used i n the c o n s t r u c t i o n of the E s t i m a t i o n Program. Actirhj  SX«15.0+(I*0.»5*LENGTH)*N SP-4.6*4.5*N SS-(6.65*0.43«WTPL*0.1•LENGTH)*N  MEASURED GLOBAL TIMES  ESTIMATED TIMES  hfonnatiee Spton Ditabase  BP 1 W920X253 0.0 2 W760X147 0.0 3 W610X140 0.0 0.0 4 W530X109 5 W460X74 0.0 0.0 67 W360X79 W250X89 0.0 e L125X125X10 0.0 9 L90X90X10 0.0 10 L90X90X10 0.0 0.0 1 1 L75X75X10 12 PL50X720 256.2 130.0 13 PL14X3860 PL30X600 303.8 14 15 PL12X1500 150.7  FEEDBACK TO ESTIMATION FUNCTIONS Not developed i n this thesis.  BS  189. 2  235. 4 0.0 153. 4 0.0 0.0 209. 4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  DP. 0.0 0.0 772.0 0.0 254.0 0.0 1236.0 0.0 0.0 0.0 0.0 188.0 132.4 120.0 109.6  PU 282. 1 450. 4 0.0 241. 3 0.0 269. 5 0.0 180. 4 110. 5 102. 3 116. 8 0.0 0.0 0.0 0.0  SA  SS  0 0 0 0 0 0 0 0 0 0 1359 2 0. 0 0 0 0 0 393 4 0 0 1 155 8 0 0 0 0 63 0 0 0 1 41 0 0 46 2 0 0 72 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0  1 2 3  516  4  1466 267 166 163 208 472 275  3 7 8  4 2131 2 4 434 2 5 647 4 6 1452 3 7  8 9 10 11 12 13 14  15  754  0) M 1 )m 2) m 3) m 4) m 5) « 6)  0.0 1.10 1.10 1 .00 1.10 1.10 1 .00  275 4  These c o e f f i c i e n t s show how the e s t i m a t i o n program performed on the l a t e s t p r o j e c t by e v a l u a t i n g the accuracy of each i n d i v i d u a l estimation function.  Figure 12. Example Use of Multiple Linear Regression (MLR).  D  4 454 2  REGRESSION COEFFICIENTS B( B( B( B( B( B( B(  4 2  30. (4)  Multiple  linear  regression Is performed  on  the two  sets of times  to  i d e n t i f y where the estimate has been i n error. (5)  The  regression c o e f f i c i e n t s  (b) indicate which f a b r i c a t i o n operations  were consistently i n error and by how much.  While the above example has demonstrated the tremendous u t i l i t y of the MLR  program one must recognize that r e a l fabrication data i s not l i k e l y to be  as  consistent as  expected  was  at a l l times.  used  there.  Variations i n productivity  The nature of MLR  are  to  be  modelling i s such that, i f given  enough data, these variations can be e s s e n t i a l l y  "ironed out" so that the  best values for the estimation parameters can be obtained.  The  degree of  certainty associated with a set of regression c o e f f i c i e n t s can be measured by examining statistics  the "goodness of f i t " provided by the given data. can  provide  an  effective  means for i d e n t i f y i n g  U t i l i z a t i o n of data  which i s  inconsistent with previous experience, and f o r locating problem areas i n the fabrication process.  3.2.4  MLR - Design and Theory The theory behind multiple l i n e a r regression i s similar to least squares  fitting of  for two dimensional sets of data.  the sum  of squared  The same concepts of minimization  residuals apply except that the number of  over which the minimization i s performed  i s not limited.  dimensions  Formulation of the  problem i n matrix form can be done as follows (see Chatterjee (1977)) [2]:  31. X  01 l l  .. .  [X] =  ?  2  On  where  ' ' *  X  .. .  )  Z  X  pl  .  ?  ..  x, ... In-  l  y  Y  2  . ..  =  x pn  —  b  ...  U =  .  y n  i  —  b =  2  u n  •.. .  2  (3)  b p  [x] are the independent variables Y  are the measured data values  1)  are regression c o e f f i c i e n t s  U_  are random disturbances  x . =1 ol  f o r a l l i ( f i r s t vector simply provides for a constant value)  n  i s the number of data values  p  i s the number of contributing variables  Estimate b by minimization of the sum of squared  S(b  l  U  b ,...b ) = 0 1 P  n I 1  u i  =  = ^  Differentiation  residuals.  (4)  2  1  ( y  ±  yields  - b  Q  -  V  l  ±  - b x 2  2 1  - ... - b ^ )  the following set of linear  (5)  2  equations  to be  solved: S  S  ll l b  21 l b  +  +  S  S  12 2  +  22 2  +  b  b  S  S  13 3  +  23 3  +  b  b  ' ' ' •  ' V " yi ' V f - y2 +  S  p  +  =  S ,b, + S „b_ + S -b + . . . + S b pi 1 p2 2 p3 3 pp p 0  where  S^ » ±  n £  (x  i k  _ - x ^ x . ^ - x^)  S  =  i,j=  <> 6  S yp  l,2,...p  (7)  32. i  S , = E (y, - y)(x., - x.) yi , , k ik i k=l w  J / v  = 1,2,...p  n E x ik k=l  (8)  (9)  n  k=l *  (10)  Solution of these equations i s done by inversion of [S].  [S] b  =  S  (11)  (pxp)(pxl) = (pxl)  providing the regression c o e f f i c i e n t s b.  b = [S]" S  (12)  1  (pxl) = (pxp)(pxl)  b_ 0  =  y - b.x- - b x. - b_x_ - . . . - b x •' 11 2 2 3 3 pp 0  (13)  The f i n a l estimate i s then:  y  4  =  b  Q  + b  l  X  + b x  u  2  2 i  + b x 3  3 i  +•  •  .+b„x p pi  (14)  With observed residuals U  i  =  y  i" i y  (15)  33. Application  of MLR  theory  to the problem  of assessing estimation  parameters  requires an examination  of the fabrication  process.  Standard  estimation  equations, which come from experience, usually relate  the time  required to perform an act to some physical c h a r a c t e r i s t i c of the material (weight, length, thickness, e t c . ) . were obtained tested  similar  through  case  dependent  Since the o r i g i n a l estimation parameters  studies, i t i s l i k e l y  characteristics  that  the investigators  f o r a l l of the operations.  For  example, one property that i s used i n almost every case i s the weight of the material.  This implies that the estimation of different f a b r i c a t i o n opera-  tions are somewhat dependent on each other, even though i n r e a l i t y they may be  very  independent  processes.  From MLR  theory  one cannot  distinguish  between deviations i n global data f o r concurrently measured dependent v a r i ables; however, this problem can be solved by ensuring that some measurements are taken when one or the other variable i s not present. t i o n of MLR i s v a l i d only i f i t i s performed  Thus, the applica-  by one who i s f a m i l i a r with the  fabrication process, and i f s u f f i c i e n t data has been c o l l e c t e d . Development of extensive software f o r use i n this area w i l l  l i k e l y be  able to overcome this b a r r i e r , so that a userfriendly program can evaluate i n d i v i d u a l applications of MLR i n terms of v a l i d i t y and l o g i c . program would eventually provide user interface knowledge of the processes  This kind of  (I/O) that requires l i t t l e  involved In multiple l i n e a r regression, so that  data can be analyzed e f f i c i e n t l y without having to perform a set of tedious mathematical operations.  34. 3.3  Applications of Program ESTImate  3.3.1  Design Optimization As  previously  explained,  s t r u c t u r a l cost  optimization  performed i n the past due  to a lack of cost information.  the design  used some other  good  profession has  common  sense,  to  a t t a i n near  design  optimum  cost  has  not  been  During this time,  t o o l s , combined with some designs.  One  not  very  successful approach has been to calculate a minimum weight design i n order to minimize  the  material  costs.  Experienced  designers  recognize  that  cost  savings due to the p e c u l i a r i t i e s of the f a b r i c a t i o n process could be attained through design s i m p l i f i c a t i o n and r e p e t i t i o n of elements. designers,  is s t i l l  to determine the  phies which produces the minimum cost The  ESTImate  program  routine of Figure 13.  can  be  The problem facing  tradeoff point between these philosodesign.  used  as  one  module i n the  optimization  Both i n t e r a c t i v e or automated a l t e r a t i o n of components  of the structure i s possible since the design evaluation program i s independent from the s e l e c t i o n procedure. functions.  One  has  Optimization  been developed by  Leung  modules can perform various [9]  to provide  minimum  cost  designs for single large components (webs of plate g i r d e r s ) . Another routine by Moses and weight design  Goble [11] to one  curve concepts. area.  of minimum cost using dynamic programming and  a minimum learning  Monte Carlo techniques could also prove to be useful i n this  Adaptations of the above techniques can a l l be inserted i n the optimi-  zation module of Figure the  takes a d i f f e r e n t approach by converting  designer.  13,  to make a variety of improvements available to  35.  Preliminary Design Yes  Figure 13.  Design Optimization.  36. 3.3.2  Improvement of the Fabrication and Estimation  Processes  The current estimation procedures used by most designers and fabricators are based on a few independent studies made i n the past of the f a b r i c a t i o n operations, with  s l i g h t modifications r e f l e c t i n g their o v e r a l l performances  on constructed projects.  The measurement of times or costs a c t u a l l y experi-  enced i n the performance of various operations i s usually done by i n s t i g a t i n g a time study program for the process i n question. usual monitoring  This i s i n addition to the  procedures used i n conventional c o n t r o l , and i t necessarily  causes additional expenses which prevent i t from becoming regular p r a c t i c e . Implementation  of an information  system,  which  collects  data  on a  regular basis from a global l e v e l , coupled with a multiple l i n e a r regression analysis can provide a continuous The  flowchart  of the ESTImate  mechanism described  above.  basis for updating program  shown  estimation parameters.  i n Figure  8 outlines the  Estimation functions contained  i n the a c t i v i t y  database can be updated by the MLR c o e f f i c i e n t s obtained tion  of  the MLR  experienced  analysis,  without  requiring  from each applica-  the a d d i t i o n a l  expenses  by time study techniques of the past.  Additional benefits provided by the ESTImate program are: automatic bid summaries and spreadsheet progress  analysis of the bid f o r pre-construction use, and  reports and schedules  f o r use during the construction.  Potential  modules can be added to the program to provide complete integration of a l l aspects analysis  of construction  from  of the f a b r i c a t i o n  accounting  process  to shipping.  Since  real-time  i s now a v a i l a b l e , investigations may  reveal where improvement i s needed and i s possible when there i s s t i l l time to make corrections.  37. 3.3.3  Research and Education A major benefit provided by program ESTImate i s access to "experience"  which may prevalent  alleviate  some of  the preference  i n the education of engineers  for a n a l y t i c a l  [9].  The  detachment  now  application of ESTImate  could be done i n the f i n a l years of an engineering student's education i n a design project.  Rather than designing to simply meet code requirements, as  Is  so often done i n education and i n professional practice, designs can  be  evaluated i n terms of economy.  with  Recursive design w i l l  hands on experience with code requirements  improve or optimize t h e i r designs.  and  now  provide students  some Ideas of how  to  Optimization routines such as that of  Figure 13 can become common knowledge among students, and hopefully w i l l be adopted by the firms which they w i l l be employed by i n the future.  38.  4. SUMMARY 4.1  Conclusions This  thesis  educators.  provides  two  new  tools  for designers, f a b r i c a t o r s ,  and  The f i r s t i s a f l e x i b l e estimation routine, able to accept a r b i -  trary process and project data. model coupled with an  The second i s a multiple l i n e a r regression  information system which w i l l  provide the  necessary  data for the estimation routine. The  estimation program can be used  both professional practice and arbitrary  structural  f a b r i c a t i o n plant.  form  by  as a design evaluation module for  research. examining  Costs the  can be associated with an  specific  capabilities  of  the  This mechanism for comparison leads to potential applica-  tions i n design improvement and h e u r i s t i c optimization. Potential improvements i n the fabrication and estimation processes are made possible with the use  of a multiple linear regression a n a l y s i s .  required for the estimation process can now  Data  be obtained by simply analyzing  data measured by a "checkstation" approach, so that the costs associated with the fabrication process can f i n a l l y be monitored i n an e f f i c i e n t manner. Integration  of  scheduling, c o n t r o l l i n g ,  and  estimating as  offered  t h i s system provides a mechanism that can assess current p r a c t i c e , and i d e n t i f y areas of p o t e n t i a l improvement. been heavily dependent these new  on  their  past  by can  Designers and f a b r i c a t o r s , who have  experience,  could  take  advantage of  tools by u t i l i z i n g them to improve the quality and competitiveness  of structural s t e e l designs.  4.2  Extensions of This Research Possible extensions  of the research contained  i n this thesis come from  the application of program ESTImate to a variety of problems.  Three s p e c i f i c  areas are noted: (1)  Optimization of s t e e l structures, and some design guidelines showing how designers  could approach this problem, could r e a l i s t i c a l l y be attained  by adopting the algorithm of Figure 13.  This problem also involves the  development of modules which t i e i n both structural analysis and requirements,  utilizing  a format which i s compatible  with  code  the rest of  the program. (2)  A more detailed analysis of the f a b r i c a t i o n process, along with a mathematical i n v e s t i g a t i o n into the s t a t i s t i c a l accuracy provided by various formulations  of  the  multiple  linear  regression  model,  i s needed  to  assess the implementation of an information system i n an existing f a b r i cation plant. (3)  Commercial use of the programs developed thus far i s not possible without input and output routines that are s p e c i f i c a l l y designed needs of the user.  to suit the  The programs must be transported to a microcomputer  where s p e c i f i c routines can be developed that are e f f i c i e n t for use with that machine.  AO.  REFERENCES [I]  ADLARD, E.J., "Computer-Integrated  Manufacturing - I t s Application and  J u s t i f i c a t i o n , " Engineering Digest, January 198A, 16-20. [2]  BERRY, G.L., "Shop Floor Information System: Design and Implementation," Engineering Digest, March 1984, 18-20.  [3]  CHATTERJEE, S., "Regression Analysis by Example," John Wiley  & Sons,  Inc., 1977. [4]  CISC, "A Project Analysis Approach to Building," 1980.  [5]  DE GARMO, E.P., "Materials and Processes i n Manufacturing," 5th Ed., New York:  [6]  MacMillan Publishing Co., Inc., 1979.  DONNELLY, J.A., "Determining  the Cost  of Welded Joints,"  Engineering  Journal, AISC, October 1968, 146-147. [7]  FORDE, B., LEUNG, Y.C. and STIEMER, S.F., "Computer-Aided Design Evaluation  of Steel  International  Structures," Presented Association  at the 12th Congress  f o r Bridge  and  Structural  of the  Engineering,  Vancouver, 1984. [8]  HAYES, R.H. and ABERNATHY, W.J., "Managing Our Way to Economic Decline," Harvard Business Review, 58 July/August 1980, 67-77.  [9]  LEUNG, Y.C. "A Contribution to Computer Aided Design Evaluation of Steel Structures", M.A.Sc. Thesis, University of B r i t i s h Columbia, 1984.  [10] MACGINLEY, T.J., "Steel Britain:  Strutures - P r a c t i c a l  Design  Studies", Great  E. & F.N. Spon Ltd., 1981, 6-9.  [II] MOSES, F. and GOBLE, G., "Minimum Cost Structures by Dynamic Programming," AISC Engineering Journal, July 1970.  41.  [12] NIXON, D.,  "Estimating  the Cost  Vol. 1, No. 2, December 1974, [13] O'BRIEN, J.A.,  "Computers i n Business  [14] PEURIFOY, R.L.,  "Estimating  McGraw-Hill Book Company,  L.A.,  Conference  Construction Costs,"  on  Walker Company, HOERNER, S., for  New  York:  1975.  "Structural Design  [17] SHOEMAKER, M.M.,  Tool  3rd Ed.,  "Cost Optimization of a Structural Roof System," M.A.Sc.  Pittsburgh, 1960,  [18] VON  Management", 3rd Ed., Homewood,  1982.  Thesis, University of B r i t i s h Columbia, [16] SCHMIT,  Steel Buildings," C.J.C.E.,  150-157.  I l l i n o i s : Richard D. Irwin, Inc.,  [15] RUSSELL, A.D.,  of Small  Electronic  by  1970. Systematic  Computation,  ASCE,  Synthesis," Structural  Second  Division,  105-132. "The Building Estimator's Reference Book," The Frank R. 1977. "The  Derivative Tensor of the S t i f f n e s s Matrix as a  Optimization,"  Optimization  Methods  in  Euromech-Colloquium 164, University of Siegen, 1982,  Structural Design, 84-90.  APPENDIX  USER'S  • ••  G U I D E  • •  • •  •• • ••  • •  V E R S I O N  1.0  m a t e  A U G  by Bruce  Forde  1984  ESTImate USER'S G U I D E  43.  Table of Contents 1. I N T R O D U C T I O N  2. A P P L I C A T I O N S  3. U S E a)  O F  T H E  Program  b)  44  of ESTImate  P R O G R A M  execution  ESTImate  46  Operations  46  i.  Editing with  the  ii.  C r e a t i n g the  Estimate  iii.  Estimating  iv.  Multiple  c)  Data  45  Variable  Linear  List  Editor  Subroutine  File  Regression  Files  53  i.  Descriptions  ii.  Examples  4. E S T I m a t e  (VLE)  C O M M A N D  5. S U M M A R I E S  L A N G U A G E  of ESTImate  66  S U B R O U T I N E S  a)  C O M P A R E  b)  C R E A T E  67  c)  D I S P L A Y  67  d)  E Q U A T I O N  68  e)  E S T I M A T E  69  f)  F I L E S T A T  69  g)  G E T S U B  69  67  70  h)  M L R  i)  R E A D  70  j)  STEP  71  k)  V L E  72  6. F I L E  N A M E  7. E R R O R 8. D A T A  I/O  73  M E S S A G E S  74  F O R M A T  a)  Activity  b)  Estimate  Database  c)  Information  d)  M L R  e)  Quantity  f)  Variable  78 (AD)  Database  79  (ED)  System  8  Database  (ISD)  Output ( M L R ) Database List  Editor  0  31 32  (QD) Files  33 (VLE)  34  ESTImate  1.  Forde  p r o g r a m was  during  estimation  in the  Use  of  minor  throughout  Read  Try  section  to  data (3)  5 to code  program  and  cornmand  of your  source  logic  use  own  in a  used  by  use  use of  detail  to  provide  structural  in  the  and  u s e d is  Equipment  -  o f this guide  the  C o l u m b i a by  Bruce  a  flexible  designers  and  costs.  language  Digital  was  both  development  The  for  of British  program  explained in  user's  construction  F O R T R A N  V A X  is  following  11  guide,  are  placed  with  77  some  environment  to:  program. Then  read  section  4  on  the  language. results to  of  of ESTImate,  expansion and  for  and  error  ensures  the  section  primarily  operation  be  code.  (refer  8 are  approach the  the  the  program  3 regarding  duplicate  Sections  This  is  suggested procedure  ESTImate (2)  for  University  o f construction  the the  the of  could  program  regarding  at  intent  estimation  adaptations  The  The  which  the  comments  strategically  developed  1984.  routine  fabricators  (1)  G U I D E  I N T R O D U C T I O N This  and  U S E R ' S  examples 8 on  those are  given  data  who  in  format  wish  included  to  to  section if  3c  using  input  necessary).  make provide  alterations a  basis  to  the  for  diagnosis.  that  of program  the  user  ESTImate.  will  become  familiar  with  both  the  ESTImate USER'S G U I D E  2. APPLICATIONS O F ESTImate There are two intended users of ESTImate: (a) Contractors  - project estimating - project scheduling - process improvement  (b) Designers  - project estimating - design evaluation/optimization  The  largest  benefits  contractors, w h o w o u l d since  it  provides  operations. of  reducing  contractors  Design costs  of  their  established.  evaluation and  have  projects;  u s e it as a t o o l  a mechanism  can benefit  Designers  offered  b y contractors  however,  profit  the resulting  a potential  ESTImate  in estimating  to monitor  improving from  by  are  and  a n d evaluate  design  projects,  the efficiency  of  an effective  that  both  to  their means  owners  and  improvements.  use o f E S T I m a t e  design/construction  so  available  scheduling  m a y also provide margins  those  in the evaluation/  optimization  interaction  first  must  be  ESTImate USER'S GUIDE  .46.  3. U S E O F T H E P R O G R A M (a) Program Execution To under  invoke  the  program ESTImate  on  a  Digital  operating system, simply  VMS  Equipment  V A X ,  which  is  running  type:  RUN EST! This  can  be  command set  of  abbreviated  file  (see  interactive  user can  create  multiple  V M S  a  user specified c o m m a n d by  operating system  c o m m a n d s are  and  set o f e s t i m a t i o n with  the  to  use a n  equations  linear  available  guide).  (see  analysing  data  ESTImate  ESTImate  estimation program, or by  Once  can  making is  of a  V M S  running,  c o m m a n d language).  evaluate  measured by  use  an  the  accuracy  information  a  The  of  a  system  regression.  (b) ESTImate Operations The  flowchart  Rectangular contained arrows.  of  Figure  boxes indicate subroutines.  Detailed  subroutines are  Further  in  the  V L E  terms  construction  ESTImate of  the  (since  V L E  the  List  utilizes V L E  can  are:  lower  be  with  has  any  the  D E S C  for  I/O  refer  are  operations  list  is  addition  can  be  self  indicated  and  for  can  several case with  length  SAVE,  default  to interactive  these)  to a  LIST,  by  upper  for  up  to  the  using  individual  (VLE)  exception of D E S C  equivalent  shapes  ESTImate.  5.  only  support  D E S C ,  used  mechanism  of  characters  D E L ,  ovular  of program  a c c o m p a n y i n g descriptions can  In  I/O  should consist  A D D ,  case  name. as a n  the  structure  ESTImate  Editor  o f lists w i t h  variable  E Q U A T I O N  commands, the  the  variables  description in  of  subroutine. O r g a n i z a t i o n o f the  of  the  subroutines  sections 3 a n d  i. Editing with the Variable The  outlines  datafiles, while  siimmaries given  1  be  alphabetically use  abbreviated D E S  and to one or  program Names  characters length  available  with in  operations.  maximum  STOP,  to  of VLE,  internal  o f 40. T h e  abbreviated  done  alphanumeric a  SORT,  be  of  10.  The  operations HELP. letter  des  to  AH  of  including prevent  ESTImate USER'S GUIDE  funct  VARIABLE  VARIABLE DATABASE - Vartable I 151 - dsscriptions  ACTIVITY DATABASE E i t i M t i o n  10m  47  INPUT  -  input  i  C  I  SUBROUTINE CREATE  -GET  (  ESTIMATE Esttiaatad for each  (  OUANTITr  Physical  SUBROUTINE 01NPUI  -  \  I  REFERENCE  to E s t I M t a  -  J</  -  C  'I  I  SUBROUTINE IS1NPUT  I N T E R A C T IVE INPUT (by Shop)  V I  INFORMATION SYSTEM OAT*B*S£ Actual Times/Costs  DATABASE Timas/Costs activity  -  /  V  I  SUBROUTINE  INTERACTIVE INPUT Est m o t o r I  Iby  INFORMATION S»STEM INPUT C O N T R O L - Input 'Orwt  proparttes  - A c t I v l t y •athOCJS  READFSTI  \  rr r  DATABASE  sue -  SUBROUTINE I ST IHA T £  RIADE5TI  -  1i  I  SUBROUTINE  J I  ESTIMATE SUBROUTINE f HE  V A R I A B L E OUTPUT CONTROL -output variables  proper tlas  (  VLE  QUANTI!r INPUT C H N R O l Input format  DATABASE  - Sac tion  descriptions  VIE  1  CISC  CONTROL  - ttipi.it var m b t e s  N  f / I  N S U B P O U T ] NE SCHEDULE  SCHEDULE OUTPUT - Progress rtporT C Scneojias  Figure 1. Program ESTImate Flowchart.  \ J  ESTImate USER'S G U I D E  confusion lower  with  level  name,  of  Output  or  range  should  Control  command  D E L  command. D E L E T E  commands, which  number,  V L E  the  of  be  in  this  case  is the  are  only  used to  V L E  command  facilitate  that  deletion  includes  by  numbers.  used  files. F u r t h e r  language  48.  to  establish  the  descriptions o f  Variable  the  Database  commands  and  are  the  given  Variable in  the  summary,  ii. Creating the Estimate Subroutine File Program the  ESTImate  set o f e q u a t i o n s  could the  be  written  correct  ESTImate. coding  must  Rather  To level  in  employed  who  o f the  that  in  user  than  to create  a  subroutine  estimation process.  with to to  program  writes the an  the  included  f o r c i n g the  simply  user  is f a m i l i a r  were  rest  rather  the  F O R T R A N , make  subroutine  This  i f it  was  create  the  ESTImate  " C R E A T E "  the  very  with  the  may for  require  the  input  file  is p r e p a r e d  with  having  to  understand  the  extensive  user.  This  a  of  set  inner  (or  an  estimate  subroutine  command  file  language  abbreviation)  Activity Database Filename Variable Database Filename  will  prompts appear  to  ?  (the  the  user  prompt  instigate  <enter ?  answer  is  "Y"  then  determine  the  the  following  Estimate Subroutine Filename  your  < enter  Do you want the output sent to a the  rest  all  of  F O R T R A N  research).  way  the  workings  the  should  must  be  be  C R E A T E  ?  your  direction  file  ?  (Y/N)  prompt  <enter  of  >  output: < Y  or  appears:  your  filename>  the  subroutine  filename> filename  in  "EST!:").  ESTlrCREATE  If  that  user  equations, in  below:  Further  ensured  with  familiar  contains  subroutine  it c o m p a t i b l e  become (which  which  of  a the  program.  o f the  shown  the  ensure  format,  ESTImate  one  than  ESTImate  only  defined  by  be  statements  used  Program  to  requires  N >  first The  word  prompts  as  a  49.  ESTImate USER'S G U I D E  The is  filename advisable  rather  than  errors.  If  (the  most  handling likely  chosen must to  first  to a •an  to  display  file)  result  i f the  are  in  where  the the  to  the  input  occurs, reference  errors  identify  non-existing since  the  to see  error  likely  be  activity error  screen  files  will  ESTImate  be  (by  are  not  as to w h e r e  See  overwrite.  c h o o s i n g this  processed  made  database).  will  the  section  occurred ( R E A D A C T ,  option  without it  It  any  originated on  error  E Q U A T I O N ,  and  STEP  candidates).  ACTIVITY DATABASE Estimation  VARIABLE DATABASE Variable List descriptions  runctiens  SUBROUTINE CREATE  ESTIMATE SUBROUTINE FILE  Figure  2. C r e a t i n g t h e  Estimate  Subroutine  File.  ESTImate USER'S GUIDE  fax  5  0  >  Estimating  Before used  to  compile  This  requires  has  to  estimate.  the estimation and link  about  be done  routine  can be invoked  the new  30 seconds o n a  once  T h e terminal  for each input  estimate V A X  G E T S U B  subroutine  file  command  to the m a i n  11/730 ; however,  n e w activity  will  the  this  database, not every  must  program.  procedure  time  be  y o u want  only to  be:  ESTLGETSUB Estimate Subroutine File  ?  < enter  your  filename  >  ESTIMATE SUBROUTINE FILE  QUANTITY DATABASE  VARIABLE OUTPUT CONTROL Output  Physical Activity  Variables  SUBROUTINE ESTIMATE  J  ESTIMATE DATABASE E s t l a s t e d Ti«es / C o s t s l o r each a c t i v i t y  Figure 3. Estimating.  Properties atethods  ESTImate USER'S GUIDE  Once the estimate subroutine file command  will  result  in  prompts  has  for  51.  been  linked,  using  the  a quantity database and an  E S T I M A T E output  file.  Reading from this data may result in some errors as described i n the section o n error  messages,  formats.  Output  with  the  correct  is controlled  by  format  the  being given  V O C (Variable  described in the previous section. The terminal  input  the section  in  Output  on  Control)  data  file  as  the E S T I m a t e p r o g r a m  will  will b e :  ESTL ESTIMATE Quantity Database Filename ? < enter your  filename  Further Prompts appear to determine the direction  of  Do you want the output sent to a file ? ( Y / N ) If  the answer is  " Y "  then  a  further  Estimate Database Filename The  output  should  file  not overwrite an existing  ?  prompt  >  output:  < Y or N >  appears:  < enter  your  be a non-existing  file  filename since  >  file,  iv. Multiple Linear Regression To  examine  database  combined  the  (containing  accuracy  of a  the estimated  w i t h the actual  times  as  set times  of  estimation  calculated  measured  by multiple linear regression. To invoke  equations,  by the estimation  by an information  this  an  analysis,  input  equations)  system  will be:  ESTL: MLR Estimate Database Filename ? <enter  filename>  your  Information System Database Filename ? <enter Further prompts follow to direct the  flow  of  " Y "  then the following  M L R Output Filename < enter  prompt  your  filename>  output:  Do you want the output sent to a file ? ( Y / N ) If the answer is  your  <y  appears:  filename  >  or N >  estimate  are  analysed  ESTImate  The  U S E R ' S  G U I D E  52.  resulting output is a datafile which contains a set of regression  coefficients which could be used to update the current set of estimation functions. This routine calls the COMPARE function which checks to see if the two input files are compatible. The definition of compatible is that every item name and description must be identical to its counterpart on the other list, and both lists must have the same number of items.  ESTIMATE DATABASE  INFORMATION SYSTEM DATABASE  E s t i a a t e d Tiaas / Costs for aach a c t i v i t y  Actual Tiaes / Costs  r  ^^S^  I SUBROUTINE MLR  MLR COEFFICIENTS Used t o update A c t i v i t y Database  Figure  4.  Multiple  Linear  Regression.  53.  ESTImate USER'S G U I D E  (c) Data Files Presented  the  datafiles  on  following pages  the  used  in  ESTImate,  and  are:  then  first  a  a list o f b r i e f d e s c r i p t i o n s  list o f  of  examples.  L Descriptions Activity Database The are in  activity  to the  be  database  used  data  by  format  to  through  optimization  routines two  estimation  section  present  arbitrary,  be  the  is c o m p o s e d o f a  under  constructed  but  must  could  make  match use  the  heading  of a  the  in  the  of  routine. T h e  using  routines  set  the  code control  format  of  system future.  given  in  estimation  file w h i c h  o f this  A D  file  is  F O R M A T .  editor, but The  the  functions  can  order  activity  The  file  new  is  is  Future  input  order  at  data  activities  database.  organizes the  described  accept  of  which  matching  the  files.  Variable Database  in  a  The  variables  V L E  format  section  on  used All  editing  in  the  I/O  with  is  the  estimation  done  V L E  by  the  process are V L E  stored  routine  alphabetically  and  is  here  described  in  the  editor.  Estimate Subroutine File The  ESF  estimation  is a  F O R T R A N  functions  in  a  77  equivalent  standard  of  program  the  activity  with  IF  database, T H E N  describing  and  the  assignment  statements.  Quantity Database The methods similar is n o t yet the  quantity  pertaining  to the done  been data  database to  material  by  a particular  bills used  ESTImate  devised. T h e format  is c o m p o s e d o f all  section.  version  format  by  project most  data  (or  properties  portion  of a  contractors. Input  1.00, s i n c e  o f the  physical  an  must  efficient follow  of  input that  and  project). this  activity This  database  mechanism  given  by  is  Q D  has  not  F O R M A T  in  ESTImate USER'S  54.  G U I D E  Variable Output Control File The with  variables  the  editor  total To  format  by simply  statement  to be output that  they  replacing the variable  T o get a F6.1 output,  an  13  output  routine  are to be displayed  field o f 6 digits (including get  b y the estimation  in. This  description with  is d o n e  with  here  1 digit  "6.1". after  use " 3 " to produce a n integer  along  using the V L £  a F O R T R A N  the description s h o u l d b e the decimal),  simply  are stored  F  This  or I  format  means  a  the decimal. with  a maximum  of 3  digits.  Estimate Database Estimated  times/costs  corresponding file if  is c o n t r o l l e d desired  IS, by  mark  each  numbers  and  activity  are  summarized  descriptions. C h o i c e  by V O C (below), a n d should be such  (if all estimated  M L R E D  for  times  can be performed  F O R M A T  are output, then  i n the data  format  with  o f the variables  that M L R  if times  o n the combination).  here  can be  are recorded  T h e format  o f this  the  in  this  performed  b y the file  is  given  field  are  section.  Information System Database The stored  times/costs  here  format  with  of  measured  by the information  the corresponding  the  datafile  must  item  be  mark  system  in the shop  numbers  that given  by ISD  and  or  descriptions.  F O R M A T  The  in the data  format  section.  M L R Output File This  file  coefficients current requires user  contains  the regression  can be used  times  o r costs actually  some  to devise  to update  judgement this  as  subroutine.  coefficients  the activity  measured  from  database  M L R  analysis.  so that  it r e f l e c t s  b y the information  to the m e t h o d  o f alteration,  system. it is left  Since  These the this  u p to the  55.  ESTImate USER'S G U I D E  it Example Data Files  The following output is from the example mentioned in the M.A.Sc. Thesis of Bruce Forde: Five operations were investigated, and for clarity of use  in the  ESTImate program these were broken down into seven groups (due to the nature of the equations used). Information was collected at the entrance and exit from the part of shop where these operations take place. Refer to the above reference for more information on the choice of the example and the use of the results.  ESTImate USER'S G U I D E  56.  Activity Database The following is a datafile which represents the estimation functions used for the estimation of seven fabrication operations. The format is of type A D (section 8a).  shearing  of plates  octivity description> o c t i v i t y number> <a s t e p f u n c t i o n i n t e r m s o f "PCWT" <range ( u n i t s a r e t h o s e o f "PCWT") <equation a p p l i e d f o r t h e above r a n g e o f t h e v a r i a b l e "PCWT" >  STEP/PCWT/3 0,25 SP=4.6+4.5*N 25,75 SP=6.32+7.2*N 75,10000 SP=6.59+9.7*N RETURN <end o f a c t i v i t y > shearing of angles 2 STEP/LENGTH/2 0,15 SA=15.0+(1+0.15*LENGTH)*N 15,1000 SA=(52+0.3*LENGTH)*N RETURN sawing o f shapes 3 SS =(6.85+0.43*WTPL+0.1*LENGTH)*N RETURN b u r n i n g o f shapes 4 BS=(1.23+0.045*WTPL)*N RETURN burning of p l a t e s 5 BP= 12.0+(4+(2.0 + 0.03*THICK)*BLENGTH/1 0 0 0 ) * N RETURN drilling 6 DR=(5+(1,0+0.05*THICK)*NHOLES)*N RETURN punching 7 PU=5.78+(0.42+0.9*NHOLES)*N RETURN END <end o f a l l a c t i v i t i e s >  ESTImate USER'S GUIDE  Variable Database  (input using VLE)  This is all set up for you by the V L E editor by simply running ESTImate and invoking the editor. The list is internally alphabetized so the input order does not matter.  NUMVAR 1 2 3 4  5 6  7 8  9 10 1 1 12 13 14  14 BLENGTH BP BS DR LENGTH N NHOLES PCWT PU SA SP SS THICK WTPL  b u r n l e n g t h (mm) time f o r burning p l a t e s time f o r b u r n i n g shapes time f o r d r i l l i n g member l e n g t h number o f i d e n t i c a l members number o f h o l e s pieceweight time f o r punching time f o r shearing angles time f o r s h e a r i n g p l a t e s time f o r sawing shapes member t h i c k n e s s weight perl e n g t h  ESTImate USER'S GUIDE The Estimate Subroutine File  (output of subroutine CREATE)  The following is the program written by the create subroutine using the activity database and the variable list The names of the variables in the equations (as input by the user) are translated into an equivalent element in the V-vector, having the corresponding number from the variable list The step functions are converted to F O R T R A N IF T H E N statements containing the values of the range given by the user. The lower number is contained in the range while the upper number is the upper bound.  C****************************************************************** C E S T I M A T E S U B R O U T I N E F I L E C <filename> C****************************************************************** SUBROUTINE ESF(V,NAME,NVAR,ACTCODE,NACT) C REAL V ( 1 4 ) INTEGER NVAR,ACTCODE(*),NACT CHARACTER*10 NAME(*) NVAR= 14 NAME( 1) = 'BLENGTH NAME( 2 ) = 'BP NAME( 3) = 'BS NAME( 4) ' DR NAME( 5) = 'LENGTH NAME( 6 ) = ' N NAME( 7 ) = 'NHOLES NAME( 8 ) ' PCWT NAME( 9) = 'PU NAME(10) = •SA NAME(11) = 'SP NAME(12) = 'SS NAME(13) = 'THICK ' WTPL NAME(14) ***************************************************************** shearing of plates 1).EQ.1)THEN IF(ACTCODE( I F ( V ( 8 ) . G E .O.AND.V( 8 K L T . 2 5 ) 1 V ( 1 1 ) = 4 . 6+4 .5*V( 6) I F ( V ( 8 ) .GE .25.AND.V( B J . L T . 7 5 ) 1 V ( 1 1 ) = 6 . 32+7.2*V( 6) I F ( V ( 8 ) .GE .75.AND.V( 8 ) . L T . 1 0 0 0 0 ) 1 V ( 1 1 ) = 6 . 59+9.7*V( 6) END I F  ESTImate USER'S G U I D E  C  59.  :********  shearing of angles  C I F ( A C T C O D E ( I F ( V (  2 ) . E Q . 1 ) T H E N  5 ) . G E . 0 . A N D . V (  1 V ( 1 0 ) = 1 5 . 0 + ( 1 + 0 . 1 5 * V ( I F ( V (  5 ) ) * V (  5 ) . G E . 1 5 . A N D . V (  1 V ( 1 0 ) = ( 5 2 + 0 . 3 * V ( E N D  5 ) . L T . 1 5 ) 6 )  5 ) . L T . 1 0 0 0 )  5 ) ) * V (  6 )  I F  C  C  sawing  o f shapes  C I F ( A C T C O D E (  3 ) . E Q . 1 ) T H E N  V ( 1 2 ) = ( 6 . 8 5 + 0 . 4 3 * V ( 1 4 ) + 0 . 1 * V ( E N D  5 ) ) * V (  6)  I F  C  C b u r n i n g o f shapes C I F ( A C T C O D E ( V (  4 ) . E Q . 1 ) T H E N  3 ) = ( 1 . 2 3 + 0 . 0 4 5 * V ( 1 4 ) ) * V (  E N D  6 )  I F  c  C burning of plates C I F ( A C T C O D E ( V (  5 ) . E Q . 1 ) T H E N  2 ) = 1 2 . 0 + ( 4 + ( 2 . 0 + 0 . 0 3 * V ( l 3 ) ) * V (  E N D  1 ) / 1 0 0 0 ) * V (  6 )  I F  ----  c  C  drilling  C I F ( A C T C O D E ( V (  6 ) . E Q . 1 ) T H E N  4 ) = ( 5 + ( 1 . 0 + 0 . 0 5 * V ( 1 3 ) ) * V (  E N D  7 ) ) * V (  6)  I F  C  ( 2 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *  C  punching  _  c  I F ( A C T C O D E ( V (  7 ) . E Q . 1 ) T H E N  9 ) = 5 . 7 8 + ( 0 . 4 2 + 0 . 9 * V (  E N D  I F  C N A C T = R E T U R N E N D  7  7 ) ) * V (  6 )  60.  ESTImate USER'S GUIDE Quantity Database  (input using the VMS editor)  The following is the quantity database (structure database, material bill, etc.) to be estimated using the chosen estimation functions and program given above. The format follows that given by the  Q D format of section 8e.  PROJECT 501 M a t e r i a l b i l l 12  < p r o j e c t name> <quantity database d e s c r i p t i o n >  (A3,X,A13)  <format of m a r k , d e s c r i p t i o n >  7 -number  of p o t e n t i a l  activities  (X,7I1) 7 -number  <format o f a c t i v i t y of input  6,5,14,8,1,7,13  code>  variables  -order of input  *  variables  <format o f v a r i a b l e s >  14 -number  of variables  (T10,A10,T25,A40) 1 BLENGTH 2 BP BS 3 DR 4 LENGTH 5 N 6 7 NHOLES PCWT 8 PU 9 SA 10 1 1 SP 12 SS THICK 13 WTPL 14 1 W920X253 0001001 15.0 12.0 253.0 2 W760X147 0001001 30 8.2 147.0 3 W610X140 0010010 20 9.1 140.0  i n database<format o f v a r i a b l e l i s t > b u r n l e n g t h (mm) time f o r burning p l a t e s time f o r b u r n i n g shapes time f o r d r i l l i n g member l e n g t h number o f i d e n t i c a l members number o f h o l e s pieceweight time f o r punching time f o r s h e a r i n g angles time f o r shearing p l a t e s time f o r sawing shapes member t h i c k n e s s weight p e r l e n g t h <mark,description> o c t i v i t y code> 0.0 920.0 20.0 28.0 0.0  760.0  16  17.0  0.0  0.0  16  22.0  61.  ESTImate USER'S GUIDE  4 W530X109 0001001 25 8.0 109.0 5 W460X74 0010010 106.7 74.0 6 W360X79 0010001 28 4.6 79.0 7 W250X89 0001010 40 6.0 89.0 8 L125X125X10 0100001 30 4.0 0.0 9 L90X90X10 0100001 18 3.0 0.0 10 L 9 0 X 9 0 X 1 0 0100001 24 2.0 0.0 11 L 7 5 X 7 5 X 1 0 0100001 50 1.0 0.0 12 P L 5 0 X 7 2 0 0000110 4 15.0 0.0 13 P L 1 4 X 3 8 6 0 0000110 2 15.0 0.0 14 P L 3 0 X 6 0 0 0000110 8 10.0 0.0 15 P L 1 2 X 1 5 0 0 0000110 4 10.0 0.0 E N D  0.0  530.0  10  19.0  0.0  0.0  12  14.0  0.0  0.0  10  0.0  0.0  250.0  14  17.0  0.0  0.0  6  0.0  0.0  0.0  6  0.0  0.0  0.0  4  0.0  0.0  0.0  2  0.0  4212.0  16440.0  12  50.0  6322.7  22720.0  36  14.0  1404.0  11200.0  4  30.0  1404.0  13000.0  14  12.0  ESTImate USER'S G U I D E  Variable Output Control  (input using VLE)  The list of output variables is contained in the V O C file as well as their output formats. In this case, all variables to do with times were output so that M L R could be performed on the operations in question. The formats used are F8.1 as discussed in the file description for V O C . N U M V A R  1 2 3 4 5 6  6 B P BS D R P U S A S S  8.1 8.1 8.1 8.1 8.1 8.1  ESTImate USER'S GUIDE Estimate Database  63.  (output by subroutine ESTIMATE)  This is the times or costs as estimated by the estimation subroutine. The input format is as given in the ED FORMAT in section 8b and is very similar to the ISD FORMAT of the next page. PROJECT 501 M a t e r i a l b i l l 12 6 Output v a r i a b l e s (T10,A10,T25,A40) 1 2 3 4 5 6  BP BS DR PU SA SS  time time time time time time  f o r burning plates f o r b u r n i n g shapes for drilling f o r punching f o r shearing angles f o r sawing shapes  (X,A3,X,A13,F8.1,F8.1,F8.1,F8.1,F8.1,F8.1)  1 W920X253 2 W760X147 3 W610X140 4 W530X109 5 W460X74 6 W360X79 7 W250X89 8 L125X125X10 9 L90X90X10 10 L 9 0 X 9 0 X 1 0 1 1 L75X75X10 12 PL50X720 13 P L 1 4 X 3 8 6 0 14 PL30X600 15 P L 1 2 X 1 5 0 0  B P 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 258.2 130.0 303.8 .150.7  B S 189.2 235.4 0.0 153.4 0.0 0.0 209.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0  D  P  R  U  0.0 0.0 772.0 0.0 254.0 0.0 1236.0 0.0 0.0 0.0 0.0 188.0 132.4 120.0 109.6  282. 1 450.4 0.0 241 .3 0.0 269.5 0.0 180.4 110.5 102.3 116.8 0.0 0.0 0.0 0.0  S A  S s  0.0  0.0  0.0  1359.2  0.0  393.4 1155.8  0.0 0.0 0.0 0.0  63.0 41.1 46.2 72.5 0.0  0.0 0.0 0.0  0.0  0.0  0.0 0.0  0.0  0.0 0.0 0.0 0.0 0.0  0.0  ESTImate USER'S GUIDE  Information System Database  (input using the VMS editor)  This is the times or costs as measured by the information system. The input format is as given in the ISD FORMAT of section 8c and is very similar to die ED FORMAT shown on the previous page. PROJECT 501 M a t e r i a l b i l l 12 6 Output v a r i a b l e s (T10,A10,T25,A40) 1 BP time 2 BS time 3 DR time 4 PU time 5 SA time 6 SS time (X,A3,X,A13,F8.1) 1 W920X253 518. 4 2 W760X147 754. 4 2131 . 2 3 W610X140 434. 2 4 W530X109 647. 4 5 W460X74 1452. 3 6 W360X79 7 W250X89 1466. 3 267. 7 8 L125X125X10 L 9 0 X 9 0 X 1 0 166. 8 9 163. 4 10 L 9 0 X 9 0 X 1 0 1 1 L75X75X10 208. 2 12 P L 5 0 X 7 2 0 472. 0 275. 4 13 P L 1 4 X 3 8 6 0 454. 2 1 4 PL30X600 275. 4 15 P L 1 2 X 1 5 0 0  f o r burning plates f o r b u r n i n g shapes for drilling f o r punching f o r shearing angles f o r sawing shapes  65.  ESTImate USER'S GUIDE Multiple Linear Regression  (output by subroutine MLR)  This is the set of regression coefficients that best fit the combination of times or costs estimated by the estimation subroutine and those measured by the information system. The results show the expected result that items 1,2,4 .and 5 should all be multiplied by a factor of 1.10 while the others remain unchanged. The matrix of times shown below are the reestimated times, (not the initial input from the information system) and are in good agreement with the information system. PROJECT 501 Material b i l l 6  12  (T10,A10,T25,A40) 1 BP time 2 BS time 3 DR time time 4 PU 5 SA time 6 SS time B( 0) = 0.0 B( 1) = 1.10 B( 2) = 1.10 B( 3) = 1.00 B( 4) = 1.10 B( 5) = 1.10 B( 6) = 1.00 1 189.2 0.0 0.0 2 0.0 235.4 0.0 0.0 0.0 772.0 3 0.0 0.0 4 1 53.4 5 0.0 0.0 254.0 0.0 0.0 6 0.0 7 0.0 209.4 1236.0 0.0 0.0 0.0 8 0.0 0.0 0.0 9 0.0 10 0.0 0.0 1 1 0.0 0.0 0.0 258.2 188.0 12 0.0 130.0 0.0 132.4 13 120.0 14 303.8 0.0 150.7 0.0 109.6 15  for for for for for for  282. 1 450.4 0.0 241 .3 0.0 269.5 0.0 180.4 110.5 102.3 116.8 0.0 0.0 0.0 0.0  burning plates burning shapes drilling punching shearing angles sawing shapes  0.0 0.0 518. 4 0.0 0.0 754. 4 0.0 1359.2 2131 . 2 0.0 0.0 434. 2 647. 4 0.0 393.4 0.0 1155.8 1452. 3 0.0 1 466. 3 0.0 267. 7 63.0 0.0 0.0 166. 8 41.1 46.2 0.0 163. 4 72.5 0.0 208. 2 472. 0 0.0 0.0 0.0 0.0 275. 4 0.0 0.0 454. 2 0.0 0.0 275. 4  ESTImate USER'S GUIDE  66.  4. ESTImate C O M M A N D L A N G U A G E  The ESTImate program is controlled by a command language which is set up in a hierarchical manner. Level 1 is accessed through the main program. Level 2 is accessed through the current main subroutines. Level 3 is accessed through the current lower level subroutines. In version 1.00 of ESTImate there is only one set of level 2 and 3 commands. This will be expanded in later versions where I/O commands will be essential. LEVEL 1 ESTImate Commands are: CREATE - creates a set of ESTImate subroutines GETSUB - select an estimate subroutine file ESTIMATE - use estimate subroutine ISINPUT - add input to the information system database QINPUT - add input to the quantity database MLR - perform Multiple Linear Regression VLE - edit variable lists DISPLAY - display a file STOP - return to V M S HELP - print this message L E V E L 2.1 V A R I A B L E LIST EDITOR Commands are: A D D - add a variable and its description D E L - delete one or more variables DESC - change a variable description LIST - display the variable list SAVE - save the new variable list SORT - sort alphabetically STOP - return to active mode H E L P - print this message L E V E L 3.1 DELETE Commands are: N A M E - deletes one variable by name NUM - deletes variable number i R A N G E - deletes variables numbered i to j STOP - return to V L E mode H E L P - print this message  67.  ESTImate USER'S G U I D E  5. SUMMARIES of ESTImate SUBROUTINES  SUBROUTINE COMPARE (VARLI ST 1 , VARLI S T 2 , NTJMVAR1 , N U M V A R 2 , COMMAND) V A R L I S T 1 V A R L I S T 2 N U M V A R 1  NTJMVAR2 COMMAND  -  the f i r s t l i s t of v a r i a b l e s . t h e second l i s t of v a r i a b l e s . t h e number o f v a r i a b l e s ( f r o m V A R L I S T 1 ) . t h e number o f v a r i a b l e s ( f r o m V A R L I S T 2 ) . c o n t r o l phrase: "SAME" - l i s t s a r e i d e n t i c a l . "ERR" - l i s t s a r enot i d e n t i c a l .  This subroutine match i d e n t i c a l l y  compares two v a r i a b l e l i s t s .  on t h e t w o l i s t s  including  Every  i t e m must  t h e number o f i t e m s .  SUBROUTINE CREATE A n a l y z i n g e s t i m a t i o n f u n c t i o n s from t h e a c t i v i t y the  l i s t of p o t e n t i a l  database,  v a r i a b l e s from t h e v a r i a b l e database,  with  CREATE  c o n s t r u c t s a FORTRAN p r o g r a m w h i c h e s s e n t i a l l y m a t c h e s t h e l o g i c o f the e s t i m a t i o n f u n c t i o n s .  The e s t i m a t i o n f u n c t i o n s must be i n p u t  according  initially  t o t h e AD f o r m a t  by h a n d , b u t c a n p o t e n t i a l l y be  u p d a t e d a u t o m a t i c a l l y u s i n g t h e MLR o u t p u t .  Subroutines  STEP a r e u s e d t o p e r f o r m  and subroutines  and  VLE a r e c a l l e d  the data a n a l y s i s ,  t o r e t r i e v e data  from  EQUATION a n d READACT  storage.  SUBROUTINE DISPLAY T h i s s u b r o u t i n e prompts t h e user  fora file  a c c e p t i n g t h e d e f a u l t o f .DAT a s t h e d a t a occur  i f t h e user  s p e c i f i e s a type  other  type. than  t o be d i s p l a y e d , An e r r o r may  a text f i l e .  ESTImate USER'S GUIDE  SUBROUTINE EQUATION  VARLIST VARDESC NUMVAR COMMAND INPUT OUTPUT  68.  (VARLI ST, VARDESC, NUMVAR, INPUT, OUTPUT)  -  the l i s t of v a r i a b l e s . the l i s t of v a r i a b l e d e s c r i p t i o n s . t h e number o f v a r i a b l e s . c o n t r o l p h r a s e ( i f s e t t o "STOP" t h e n V L E r e t u r n s t o M A I N or c a n be s e t t o "GO" t o become an i n t e r a c t i v e e d i t o r ) . - a character s t r i n g containing the input equation. - a c h a r a c t e r s t r i n g c o n t a i n i n g the output equation.  T h i s s u b r o u t i n e was d e v e l o p e d and a c t s a s t h e i n t e r p r e t e r  i n a t e s t p r o g r a m c a l l e d INT  of e q u a t i o n s used i n ESTImate.  Input  e q u a t i o n s a r e decomposed i n t o v a r i a b l e s , c o n s t a n t s , and o p e r a t o r s and t h e l o g i c  o f t h e e x p r e s s i o n i s t e s t e d . The r e s u l t  e q u i v a l e n t FORTRAN e q u a t i o n w r i t t e n t o OUTPUT. E a c h must c o n s i s t a string can  o f a v a r i a b l e name a n d an e q u a l s  equation  s i g n f o l l o w e d by  o f v a r i a b l e s , c o n s t a n t s a n d o p e r a t o r s . V a r i a b l e names  o n l y be made up o f n o n - n u m e r i c  EXAMPLE:  i s an  capital  letters.  TIME = 10.1 * (WEIGHT + LENGTH/1.62)  NOTATION: L = letter A , B , C , .. . , Z D = digit 1 , 2 , 3 , . . . , 0 0 = operator ( + , - , * , / ) LSIG = l a s t s i g n i f i c a n t c h a r a c t e r NDECIMAL = number o f d e c i m a l p l a c e s p e r c o n s t a n t  (maximum o f 1)  LOGIC: L can f o l l o w  L  D can f o l l o w 0 can f o l l o w  D L  0  (  L  L  (  D D  = )  0  . can f o l l o w = can f o l l o w  (  D  ( can f o l l o w ) can f o l l o w  0  ) 0  (  or blank  i f L S I G i s 0, ( o r =  or blank  i f L S I G i s 0, ( o r =  or blank  i f L S I G i s L, D o r )  or blank  i f LSIG i s 0 or (  or blank  i f L S I G i s L, D o r )  or blank i f LSIG i s 0 or ( B U T o n l y i f NDECIMAL=0 ( r e s e t ) or blank i f LSIG i s L B U T M U S T FOLLOW F I R S T V A R I A B L E  ESTImate USER'S G U I D E  69.  SUBROUTINE ESTIMATE  Once l i n k e d t o t h e e s t i m a t e  subroutine f i l e ,  this  subroutine  estimates the values of a l l v a r i a b l e s i n the v a r i a b l e output file  VOC.DAT, u s i n g t h e q u a n t i t y d a t a b a s e o f t h e u s e r ' s  Subroutines  V L E a n d READQUAN a r e c a l l e d  S U B R O U T I N E FILESTAT  to retrieve  control  choice.  data.  (NAME, TYPE, STAT)  NAME - t h e name o f t h e f i l e TYPE - t h e d a t a t y p e STAT - a l o g i c a l v a r i a b l e The c a l l  to this  subroutine should supply  t h e name o f a f i l e t o  be t e s t e d , a n d a v a l u e o f STAT ( a l o g i c a l  variable).  INPUT VALUE o f STAT  TRUE  TRUE  FALSE  FALSE  FILESTATUS  EXIST  NONEXIST  EXIST  NONEXIST  OUTPUT VALUE o f STAT  TRUE  FALSE  TRUE  FALSE  ERROR MESSAGE  NO  YES  YES  NO  The i n p u t o f a new f i l e n a m e must be done by t h e c a l l e r .  SUBROUTINE G E T S U B ( S U B F I L E )  S U B F I L E - name o f t h e e s t i m a t e  subroutine  file  T h i s s u b r o u t i n e makes u s e o f t h e VAX Run Time L i b r a r y f u n c t i o n w h i c h a c c e p t s a command t o c o m p i l e subroutine f i l e called  and l i n k t h e s e l e c t e d e s t i m a t i o n  w h i l e u s i n g E S T I m a t e . A command f i l e  GETSUB.COM i s  t o p e r f o r m t h i s a c t i o n p a s s i n g t h e name o f t h e f i l e  p a r a m e t e r . I n E S T I m a t e v e r s i o n 1.00 t h i s on a VAX 11/730.  as a  r e q u i r e s a b o u t 30 s e c o n d s  ESTImate USER'S GUIDE  70.  SUBROUTINE MLR This  subroutine  was d e v e l o p e d t o i n v e s t i g a t e t h e p o t e n t i a l u s e  of a m u l t i p l e l i n e a r estimation version of  regression  model  i n the evaluation  of t h e  o f s t r u c t u r a l s t e e l f a b r i c a t i o n c o s t s . The p r e l i m i n a r y  specifically  dealt with  a c a s e s t u d y where t h e p r o g r e s s  s p e c i f i c member t y p e s h a d b e e n m o n i t o r e d by an  system, which records checkstations  information  t h e t i m e a t w h i c h t h e members p a s s key  i n t h e f a b r i c a t i o n p l a n t . The e s t i m a t i o n  was s i m p l i f i e d  so t h a t a w o r k i n g program c o u l d  a short  of  period  process  be w i t t e n  over  time.  S u b s e q u e n t l y t h e E S T I m a t e p r o g r a m was w r i t t e n , a n d was done i n s u c h a manner s o t h a t MLR c o u l d  be a t t a c h e d  l i m i t a t i o n s o f t h e c a s e s t u d y were e l i m i n a t e d  t o i t . The o r i g i n a l so t h a t  t h e use of  MLR i s now u n r e s t r i c t e d . ESTImate v e r s i o n system data;  however, i t w i l l  t h e MLR r o u t i n e  R E A D  subroutine  1.00 h a s no i n p u t m e c h a n i s m  f o r information  a c c e p t d a t a o f ISD f o r m a t f o r u s e i n  (see s e c t i o n 8 c ) .  group  SUBROUTINE READNAME(NUMBER,NAME) SUBROUTINE READNUM(NUMBER,NAME) SUBROUTINE READACT(UNIT,ACTNUM,ACTDESC,ACTDAT,NEQUA) SUBROUTINE READQUAN(UNIT,NAME,DESC,NUMVAR,FMT, VARLIST,VARDESC,MARK,QDESC,ACTCODE,NACT,V,COMMAND) SUBROUTINE READESTI(UNIT,NAME,DESC,NUMOUT,OUTLIST,OUTDESC, NUMQUAN,MARK,QDESC,X,I,J) SUBROUTINE READIS(UNIT,NAME,DESC,NUMOUT,OUTLIST,OUTDESC, NUMQUAN,MARK,QDESC,Y,I) Each o f t h e above r o u t i n e s a r e used as mechanisms f o r collecting  data  f r o m t h e a v a i l a b l e d a t a b a s e s . See s e c t i o n s  on e r r o r m e s s a g e s a n d f i l e  organization  f o r more  information.  ESTImate USER'S GUIDE  71.  SUBROUTINE STEP (ACTDAT, POINTER, VARLI ST, NTJMVAR, STEPDAT, N) ACTDAT POINTER VARLIST NTJMVAR STEPDAT N The  -  the a c t i v i t y data l i s t . the location i n the a c t i v i t y data. the l i s t of v a r i a b l e s . t h e number o f v a r i a b l e s . t h e step f u n c t i o n data l i s t . t h e number o f s t e p f u n c t i o n e q u a t i o n s .  o r g a n i z a t i o n of a s t e p f u n c t i o n i s as f o l l o w s :  S T E P / < v a r i a b l e name>/<number o f <value1>,<value2> . . . equation . . . <value3>,<value4> . . . equation . . .  equations>  <valueN-1>,<valueN> . . . equation . . . The entire  word "STEP" i s r e a d by t h e c a l l i n g s e t of data  isstill  ACTDAT, w i t h t h e p o s i t i o n  i te x i s t s .  statements  the f i r s t  t h e b o u n d s f o r <varname> w h i c h  apply  e q u a t i o n a r e r e a d . S u b r o u t i n e EQUATION i s c a l l e d line  r e t u r n i n g t h e FORTRAN e q u a t i o n  t h e n w r i t t e n w i t h t h e I F THEN s t a t e m e n t  This procedure The  drawn f r o m V A R L I S T t o s e e  i n t h e s u b r o u t i n e s , e l s e an e r r o r message i s  read the f o l l o w i n g is  subroutine i n vector  I f i t d o e s , t h e n V ( i ) i s u s e d i n t h e I F THEN  printed. Following this to  to this  i n t h e v e c t o r m a r k e d by POINTER. The  v a r i a b l e <varname> i s f i r s t if  passed  r o u t i n e , but the  continues u n t i l the last  number o f s p a c e s  to  which  t o v e c t o r STEPDAT.  equation  i s processed.  u s e d i n STEPDAT i s r e t u r n e d i n N.  72.  ESTImate USER'S GUIDE  SUBROUTINE VLE( VARLI ST, VARDESC, NUMVAR, COMMAND, V A R F I L E ) VARLIST VARDESC NUMVAR COMMAND  -  the l i s t of v a r i a b l e s . the l i s t of v a r i a b l e d e s c r i p t i o n s . t h e number o f v a r i a b l e s . c o n t r o l phrase: "STOP" - r e t u r n s t o MAIN w i t h V A R L I S T a n d VARDESC. "GO" - VLE becomes an i n t e r a c t i v e e d i t o r . "NEW" - a new f i l e c a n be o p e n e d i n t h e e d i t o r . "ERR" - means t h a t an e r r o r o c c u r r e d i n V L E . V A R F I L E - t h e name o f t h e d a t a f i l e w h i c h c o n t a i n s t h e v a r i a b l e s . T h i s s u b r o u t i n e p e r f o r m s two 1. V A R I A B L E L I S T EDITOR - s o r t s , adds, and d e l e t e s -displays the l i s t  functions: variables  from a  list,  2.  I/O O f L I S T TO F I L E STORAGE - r e t r i e v e s v a r i a b l e l i s t s from s t o r a g e files. The f i l e i s a n i n t e r n a l l y a l p h a b e t i z e d s e q u e n t i a l  type.  ESTImate USER'S GUIDE  73.  6. FILE NAME I/O NOTE: later versions of ESTImate will provide access to bypass these prompts by stacking of the commands issued by the user. The following is a list of the prompts issued by ESTImate: C R E A T E  Activity  Database  Variable Do  Database  you  want  Estimate  the  Filename  ?  Filename output  Subroutine  ?  sent to a  Filename  file  ?  (Y/N)  ?  D I S P L A Y  File  to  The  default  If  you  then  be  displayed  want  you  ?  file  type  to  display  some  to write  the  have  is  F I L E . D A T other full  type  name.  E S T I M A T E  Quantity Do  you  Database want  Estimate  the  Filename output  Database  ?  sent  Filename  to  a  file  ?  (Y/N)  ?  GETSUB Estimate  Subroutine  Filename  ?  M L R Estimate  Database  Information Do  you  M L R  Filename  System  want  Output  the  Database  output sent  Filename  ?  ? Filename to a  file  ? ?  (Y/N)  ESTImate USER'S G U I D E  7. ERROR M E S S A G E S The following list shows the subroutines where a particular error message originates so that efficient diagnosis of any non- recoverable problem is possible. COMPARE Number  o f variables  in  listl:  Number  o f variables  in  list2:  D A T A F I L E S •  S E L E C T E D  • • • • •  A R E  E R R O R  D A T A F I L E S  I N C O M P A T I B L E  • • * • • •  S E L E C T E D  A R E  I N C O M P A T I B L E  CREATE The  names  o f the  . . . . . . . .  " V O C "  two  files  -TRY  is r e s e r v e d  for  . . . . . . . .  must be  A  G  the  -TRY  A  I  N  different  . . . . . . . .  O U T P U T A G A I N  C O N T R O L  F I L E  • • • • • • • •  DISPLAY A n  error  occurred  when  data  A n  error  occurred  when  the  The  most  likely  1. I n c o r r e c t 2. D a t a  file  with  causes  of  this  specification  unusual  was file  being was  read  O P E N E D  are:  for  type.  attributes.  EQUATION •••••• E R R O R  IN  V A R I A B L E  •••••• E R R O R  IN  C O N S T A N T  D E S I G N A T I O N  •••••• E R R O R  IN  O P E R A T O R  P L A C E M E N T  •••••• E R R O R  IN  L E F T  •••••• E R R O R  IN  R I G H T  •••••• E R R O R  IN  D E C I M A L  •••••• T O O  M A N Y  •••••• E R R O R  IN  B R A C K E T  N O T  M A N Y  •••••• V A R I A B L E  P L A C E  D E C I M A L  P L A C E M E N T  ••••••  ••••••  P L A C E M E N T  B A L A N C E D SIGNS  "'.VARNAME,'"  •••••• ••••••  ••••••  P L A C E S  SIGN  E Q U A L S  ••••••  P L A C E M E N T  B R A C K E T  E Q U A L S  •••••• B R A C K E T S •••••• T O O  N A M E  •  •••••• ••••••  D O E S  N O T  E X I S T  ESTIMATE You the  have  not  linked  E S T I M A T E  " V O C "  is r e s e r v e d  • • • • • • • •  E R R O R  T O O  solution: or  reduce  R E V I S E  the  for  M A N Y  (use  subroutine  the  V A R I A B L E  TRY  several  the  size o f the  output  D A T A F I L E  file  to  G E T S U B  command).  O U T P U T  C O N T R O L  A G A I N  V A R I A B L E S  use T H E  estimate  subroutine  files  variables.  IN  O U T P U T  LIST  ••••••  ESTImate  USER'S  G U I D E  M L R •  E R R O R  D A T A F I L E S  • • • • • •  S E L E C T E D  A R E  I N C O M P A T I B L E  R E A D A C T  •••••• E R R O R The  most  (OR  E N D  c o m m o n errors  1. A c t i v i t y  description  2. A c t i v i t y  number  3. R E T U R N See  the  "Organization  of  user the  R E A D ) are:  ••••••  missing.  missing.  or E N D  ESTImate  IN found  missing. guide  regarding  Activity  File".  R E A D E S T I  •••••• E N D  IN  •••••• E R R O R *  E N D  IN  E R R O R E N D  IN  •••••• E R R O R •••••• E N D  IN  •••••• E R R O R •••••• E N D  IN  E R R O R •  E N D  *  ••*••* See  the  IN  IN IN IN IN IN IN IN  ESTImate  "Organization  of  E  D  E  S  C  •  ••••••  N U M O U T  ••••••  FMT(l) O F  *  FMT(l)  B L A N K O F  ••••••  ••••••  B L A N K  O U T L I S T O F  •  •••*•• ••••••  O U T L I S T  ••••••  R E A D R E A D  O F  user the  M  F M T ( 2 ) •••••• O F F M T ( 2 ) *  R E A D  E R R O R  O F  O F  R E A D  A  ••••••  N U M O U T  O F  R E A D  R E A D  O F  O F  R E A D  R E A D  N  D E S C  O F  R E A D  R E A D  N A M E O F  O F  R E A D  R E A D  E R R O R  O F  R E A D  R E A D  IN  E R R O R E N D  R E A D IN  O F  guide  Estimate  X  V E C T O R  regarding File".  •••••  ESTImate USER'S GUIDE  R E A D I S  • • • • • • E N D IN R E A D O F N A M E • • • • • • • • • • • • ERROR IN R E A D O F N A M E • • • • • • • • • * • • E N D IN R E A D O F DESC • • • • • • ERROR IN R E A D O F DESC • • • • * • • • • • • • E N D IN R E A D O F N U M O U T • • • • • • ERROR IN R E A D O F N U M O U T • • • • • • • E N D IN R E A D O F FMT(1) * ERROR IN R E A D O F FMT(l) • • • • • • • • • • • • E N D IN R E A D O F B L A N K • • • • • • • ERROR IN R E A D O F B L A N K • • • • • • • E N D IN R E A D O F O U T U S T • • • • • • • • • • • • ERROR IN R E A D O F OUTLIST • • • • • • • • • • • • E N D IN R E A D O F FMT(2) •••••• • • • • • • ERROR IN R E A D O F FMT(2) • ERROR IN R E A D O F Y VECTOR • • • • • • See the ESTImate user guide regarding "Organization of the Information System File" READQUAN F R E E F O R M A T IS NOT A L L O W E D FOR M A R K or DESC E N D IN R E A D O F N A M E ERROR IN R E A D OF N A M E • • • • • • • • • • • • E N D IN R E A D O F DESC • • • • • • ERROR IN R E A D OF DESC • • • • • • • • • • • • E N D IN R E A D O F N A C T • • • • • • • • • • • • ERROR IN R E A D O F NACT • • • • • • E N D IN R E A D OF N V A R • ERROR IN R E A D OF N V A R • • • • • • E N D IN R E A D O F ORDER * ERROR IN R E A D O F ORDER • • • • • • . . . . . . E N D IN R E A D O F N U M V A R • • • • • • * ERROR IN R E A D OF N U M V A R • • • • • • E N D IN R E A D O F FMT(l) • • • • • • ERROR IN R E A D OF FMT(l) • • • • • • . . . . . . E N D IN R E A D O F FMT(2) •••••• • • • • • • ERROR IN R E A D O F FMT(2) •••••• E N D IN R E A D OF FMT(3) • • • • • • ERROR IN R E A D OF FMT(3) •••••• * E N D IN R E A D O F FMT(4) •••••• • • • • • • ERROR IN R E A D O F FMT(4) * * E N D IN R E A D O F VARIABLES) • • • • • • • • • • • • E N D IN R E A D O F V A R I A B L E ^ ) • • • • • • • • • • • • E N D IN R E A D O F M A R K * • • • • • • ERROR IN R E A D O F M A R K * . . . . . . END IN.READ OF A C T C O D E < I ) * • • • • • • ERROR IN R E A D O F ACTCODE(I) . . . . . . E N D IN R E A D O F V ( O R D E R ( I ) ) * . . . . . . ERROR IN R E A D OF V ( O R D E R ( I ) ) * See the ESTImate user guide regarding "Organization of the Quantity File".  76.  ESTImate  USER'S  G U I D E  S T E P  •••••• V A R I A B L E E R R O R E R R O R  IN  " V A R N A M E "  IN  N U M B E R  R E A D  • * T O O  M A N Y '  ••••**  E R R O R  O F  N O T  EXIST  A C T D A T  D E C I M A L IN  D O E S  ••••••  P L A C E S * *  N U M B E R  •••••  VLE Youi the  data  file  number  T H E  has  a  mistake  o f variables  A C T U A L  in  " N U M V A R "  N U M B E R  IS  N O T If  you  this  S A V E  will  Number  the  file,  be  corrected  of  blank  E R R O R  N A M E  is a l r e a d y  N A M E  is n o t  N o N O  on  variable  of  variable  variable  =  ••••••  list  list  D E L E T E D  name W E R E  If  J  of  D E L E T E D  number  is l e s s t h a n  only  VARLIST(I)  N O  V A R I A B L E S  E R R O R  No  the  FILE  WAS:  variable  E R R O R  encountered  D A T A  W E R E  V A R I A B L E S  N O T E :  on the  again  WAS:  E R R O R N o  B A D  V A R I A B L E S  E R R O R  start you.  variables  •  N O  and for  IN  in  then  R A N G E  W E R E  WAS:  variables  I,  only  is d e l e t e d .  range  N U M B E R  D E L E T E D  *  ESTImate  8.  USER'S  DATA FORMAT The  (Fig  flowchart  1) d i s p l a y s  following The  format  list  showing  the o r g a n i z a t i o n  a l l datafiles  identifies  types are given  as b e i n g  enclosed  in alphabetical format  PROCESS DATA  - A C T I V I T Y DATABASE  AD  - VARIABLE DATABASE  VLE  - MLR OUTPUT  MLR  B. PROJECT DATA - QUANTITY DATABASE  QD  - ESTIMATE DATABASE  ED  -  INFORMATION  IS  -  BID OUTPUT  N/A  -  SCHEDULE OUTPUT  N/A  SYSTEM DATABASE  C. ESTImate CONTROL  DATA  - VARIABLE INPUT CONTROL  VLE  - QUANTITY  QIC  -  INPUT CONTROL  INFORMATION  SYSTEM INPUT CONTROL  - VARIABLE OUTPUT CONTROL  o f t h e ESTImate  the type of format  ESTImate D a t a f i l e s : A.  78.  G U I D E  I SIC VLE  used  order  program  by r e c t a n g l e s . by e a c h following  The  datafile. this  page.  ESTImate  USER'S  79.  G U I D E  AD FORMAT The o r g a n i z a t i o n  of the a c t i v i t y  activity  description  ACTIVITY  NUMBER  file  i s as f o l l o w s :  . . . equations . . . RETURN  activity  description  ACTIVITY  NUMBER  . . . equations  . ..  RETURN END  F o r more i n f o r m a t i o n on t h e e q u a t i o n s , s e e EQUATION a n d STEP s u b r o u t i n e summaries. A l s o see t h e example  ESTImate  U S E R ' S  80.  G U I D E  ED FORMAT The o r g a n i z a t i o n  of the estimate  file  NAME DESC NUMOUT FMT(1) 1 2 3  Name o f q u a n t i t y f i l e - Quantity data d e s c r i p t i o n - Number o f o u t p u t v a r i a b l e s - Format o f v a r i a b l e numbers, OUTLISTO) OUTDESCO) OUTLIST(2) 0UTDESC(2) OUTLISTO) OUTDESCO)  NUMOUT  OUTLIST(NUMOUT)  FMT(2)  MARK  V(1)  See t h e e x a m p l e  V(2)  names a n d d e s c r i p t i o n s  OUTDESC(NUMOUT)  - Format of e s t i m a t e  QDESC  i s as f o l l o w s :  V(3)  in section  output  .  3cii  V(NUMOUT)  ESTImate  I S D  The  U S E R ' S  G U I D E  81.  F O R M A T  organization  NAME DESC NUMOUT FMT(1) 1  of the information  - Name o f q u a n t i t y file - Quantity data description - Number o f o u t p u t variables - Format o f v a r i a b l e numbers, OUTLIST(1) OUTDESC(1)  2  0 U T L I S T ( 2 )  OUTDESC(2)  3  OUTLIST(3)  OUTDESC(3)  NUMOUT FMT(2)  MARK  See  system  file  names  and  OUTLIST(NUMOUT) OUTDESC(NUMOUT) - Format of i n f o r m a t i o n system output  QDESC  Y(l)  t h e example  in section  3cii.  i s as f o l l o w s :  descriptions  ESTImate  USER'S  82.  G U I D E  MLR FORMAT The o r g a n i z a t i o n NAME DESC NUMOUT FMT(1) 1 2  3  NUMOUT  o f t h e MLR o u t p u t  file  i s as f o l l o w s  - Name o f q u a n t i t y f i l e -. Q u a n t i t y d a t a d e s c r i p t i o n - Number o f o u t p u t v a r i a b l e s - F o r m a t o f v a r i a b l e numbers, names a n d d e s c r i p t i o n s OUTLIST(1) OUTDESC(1)  OUTLIST(2)  OUTLIST(3)  OUTLIST(NUMOUT)  0UTDESC(2)  OUTDESC(3)  OUTDESC(NUMOUT)  B(0)  B(1 ) B(2)  B(NUMOUT) See t h e e x a m p l e  i n section  3cii.  ESTImate  Q D  U S E R ' S  G U I D E  83.  F O R M A T  The o r g a n i z a t i o n  NAME DESC FMT(1)  of the q u a n t i t y  file  i s as  follows  QUANTITY DATABASE NAME q u a n t i t y database d e s c r i p t i o n f o r m a t o f f i r s t d a t a l i n e (member mark & d e s c r i p t i o n )  NACT  number o f a c t i v i t i e s format o f second data l i n e ( t h e a c t i v i t y code) number o f i n p u t v a r i a b l e s V NVAR ORDER(nvar) o r d e r of i n p u t v a r i a b l e s i n q u a n t i t y d a t a b a s e format of t h i r d data l i n e ( t h e input var i a b l e s ) FMT(3) number of v a r i a b l e s V NUMVAR format of v a r i a b l e l i s t . FMT(4) FMT(2)  (1 ) ( 2 )  (3)  VARLIST(1)  VARDESC(1)  VARLIST(3)  VARDESC(3)  VARLIST(2)  VARDESC(2)  (NUMVAR) VARLIST(NUMVAR) VARDESC(NUMVAR)  data l i n e 1 data l i n e 2 data l i n e 3 data l i n e 1 data l i n e 2 data l i n e 3 data l i n e 1 data l i n e 2 data l i n e 3  data l i n e 1 c o n t a i n s : - Member mark, Member  description  data l i n e 2 c o n t a i n s : - A c t i v i t y method c o d e data l i n e 3 c o n t a i n s : - V e c t o r V(NVAR) f o r t h e member  data l i n e 1 data l i n e 2 data l i n e 3  N END See  data l i n e data l i n e data l i n e (control  1 2 3 f o r l o o p i n g ) must be i n t h e f i r s t  t h e e x a m p l e i n s e c t i o n 3c i i .  column !  ESTImate  USER'S  84.  G U I D E  VLE FORMAT The o r g a n i z a t i o n  o f VLE f i l e s  i s as f o l l o w s : FORTRAN  NUMBER OF VARIABLES 1 NAME(1) 2 NAME(2) 3 NAME(3)  DESC(1) DESC(2) DESC(3)  NUMVAR  DESC(NUMVARI  NAME(NUMVAR)  See t h e e x a m p l e i n s e c t i o n  format  (T10,I3)  3cii.  (T10,A10,T25,A40)  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0062467/manifest

Comment

Related Items