FISHING VESSEL OPTIMIZATION - A DESIGN TOOL by THOMAS CHARLES BOWER B.Sc, Royal Roads Military College, 1978 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ^SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF MECHANICAL ENGINEERING We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January, 1985 Â© Thomas Charles Bower, 1985 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. _ . Mechanical Engineering Department of The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date ABSTRACT Rising fuel costs and decreased catch sizes have reduced the fishing vessel owners profit margin. This has caused the owners to try to find methods that reduce the costs of their operations. In this thesis a tool which can be used by fishing vessel designers, and operators, i s developed for use at the preliminary ship design stage. It is used to determine the best fishing vessel parameters for a given operational scenario found on the West Coast of Canada. - i i -TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i i i LIST OF TABLES v LIST OF FIGURES v i ACKNOWLEDGEMENTS v i i NOMENCLATURE v i i i 1. INTRODUCTION 1 2. FISHING VESSEL DESIGN MODEL 4 2.1 Relationship Between the Ship Vector and Ship Dimensions .. 6 2.2 Limits Assigned to Components of Ship Vector 7 2.3 Equations Used for Fishing Vessel Design Model 8 2.3.1 Resistance and Power 8 2.3.2 Weight Equations 11 2.3.3 Volume Calculations 17 2.3.4 Stability 21 2.3.5 Cost Calculations 22 3. THE OPERATIONAL SCENARIO 26 4. OPTIMIZATION CRITERIA 28 4.1 Steps Involved in Evaluating the Cost Function 29 5. VALIDATION OF DESIGN PROBLEM 31 6. SENSITIVITY ANALYSIS 33 6.1 Cost Function vs. Changes in Length 34 6.2 Cost Function vs. Changes in Beam 34 - i i i -Page 6.3 Cost Function vs. Changes in Draft 35 6.4 Cost Function vs. Changes in Midship Coefficient 35 6.5 Cost Function vs. Changes in Prismatic Coefficient 35 6.6 Fuel Cost vs. Changes in Design Variables 36 7. ROUTINE FOR OPTIMIZATION 37 7.1 Results of the Optimization 38 7.2 Comparison of Optimum Vessel to "Eastward Ho" 39 7.3 Effect of Rising Fuel Costs on Optimum Design 41 7.4 Percent Fish Hold Capacity Required to Break Even 42 8. DISCUSSION OF FUTURE TRENDS IN FISHING VESSEL DESIGN 45 8.1 Recommendations for Future Work 46 REFERENCES 61 APPENDICES: A - Comparison of Steel Weight Estimates to West Coast Invoiced Steel Weights 63 B - Questionnaire Used to Discuss Future Trends in Fishing Vessels 66 C - Problems Encountered with Computer Programs 67 - iv -LIST OF TABLES Page Table I Values of d i , 11(1=1,4) 9 Table II Comparison Between Eastward Ho and Values Predicted in Design Program 31 Table III Results of Optimization for 16 m Vessel 39 Table IV Comparison of Optimized Ship to Eastward Ho 39 Table V Fueld Price Increase Effects 42 Table VI % Fish Hold Capacity Required to Break Even 44 - v -LIST OF FIGURES Page Figure 1 Simplified Waterplane Arrangement 17 Figure 2 Simplified Side View Showing Accommodation Below Waterline 20 Figure 3 Operational Scenario 27 Figure 4 Steel Weight Vs. Length Changes 48 Figure 5 West Coast Beam Vs. Depth 49 Figure 6 West Coast Beam Vs. Length 50 Figure 7 Cost Function Vs. Length 51 Figure 8 Cost Function Vs. Beam 52 Figure 9 Cost Function Vs. Draft 53 Figure 10 Cost Function Vs. Midship Coefficient 54 Figure 11 Cost Function Vs. Prismatic Coefficient 55 Figure 12 Fuel Cost Vs. Length 56 Figure 13 Fuel Cost Vs. Beam 57 Figure 14 Fuel Cost Vs. Draft 58 Figure 15 Fuel Cost Vs. Midship Coefficient 59 Figure 16 Fuel Cost Vs. Prismatic Coefficient 60 - v i -ACKNOWLEDGEMENTS I wish to express my gratitude for the time and assistance given to me by: Mr. Robert Allan Mr. B i l l Cleaver Mr. Aurthur MacLaren Mr. Gary Sigmund I would also like to thank, Eric Jonk, Otto Fung and Gary Lepp for their help i n improving my programming techniques and my understanding of the use of the computer systems at the University of British Columbia. Finally, I would like to thank Dr. Sander Calisal, my advisor, as well as Mr. R.H. Mcllwaine whose patience and ideas helped see this project through completion. This work was supported by the Department of Fisheries and Oceans. - v i i -NOMENCLATURE AM Midship Area A Waterplane Area w B Ship Beam B/T Beam Draft Ratio CCOST Construction CD Depreciation Cost CF Fuel Cost CFW Fresh Water Consumption Ci Equations of Form Used in Resistance Algorithm CI Insurance Cost CLVOL Crew Living Volume CM Midship Coefficient CMF Cost Function CMI Miscellaneous Costs (Repairs, Licence, Port fees, Wages) CMO Mortgage Cost CMY Machinery Weight Constant COUT Outfit Weight Constant CP Prismatic Coefficient CS Steel Cost CSTRUCT Hull Weight Constant CUBIC NO. LxBxD/2.834 CWA Catch Weight Availability (DISP - Z Weight Groups) CWL Parameter for Entrance Angle at Waterline (IE * L/B) CWP Waterplane Coefficient - v i i i -di Coefficients used i n Ci (Table 1) D Ship Depth at Midships DISP Ship Displacement at DWL DWL Design Waterline EGR Engine Gear Ratio EHP Effective Horsepower EO Propeller Efficiency ER Relative Rotative Efficiency FB Distance of LCB From Forward Perpendicular FBD Ship Freeboard FCR Fuel Consumption Ratio FHC Fish Hold Capacity (Vessel Vol. - Z Volume Groups) FN Froude Number (V//gL) FR Fuel Required to Complete 9 Days at Design Speed FVVOL Fuel Volume FWVOL Fresh Water Volume GM Metacentric Height GMR Required Metracentric Height i Any Integer IE Entrance Angle j Any Integer J Propeller Advance Ratio k Any Integer KQ Torque Coefficient KT Thrust Coefficient - ix -L Ship Length L/B Length Beam Ratio LCB [.5L-FB/L] * 100 LCB Longitudinal Centre of Buoyancy MACV Machinery Volume N Crew Size NU Kinematic Viscosity OPC Overall Propulsive Coefficient RF Frictional Resistance RN Reynold Number (VL/NU) RO Density of Seawater RR Residuary Resistance RT Total Resistance S Approximation to Wetted Surface Area STVOL Stores Volume t Thrust Deduction Factor T Ship Draft V Ship Speed (m/sec) VL Speed Length Ratio VS Ship Speed in Kts VESSEL VOL Ship Volume up to Main Deck VOL Ship Volume to DWL W Taylor Wake Fraction WF Fuel Weight WM Machinery Weight WO Outfit Weight - x -WS Steel Weight WWS Stores and Fresh Water Weight X Dimension Coefficient Xi Components of Ship Vector ( i = 1,6) - x i -1. CHAPTER 1 INTRODUCTION Over the past ten years owners of fishing vessels have had their profit raargines reduced because the catch yields have decreased while the cost of operations, especially fuel prices, have increased. The owners (of fishing vessels) expressed a desire; through the University of British Columbia and the Department of Fisheries and Oceans; to find, If possible, a more efficient method to conduct fishing operations. An obvious recommendation to reduce operating costs would be to decrease vessel speed, which would decrease fuel costs, but this may not be acceptable to operators in a l l cases. The main problem with speed reduc-tion i s a result of fishing seasons being limited in duration, as well as being opened In different areas at different times. The question of how to Improve economic efficiency for the Pacific Coast fisherman was looked at in two stages. The f i r s t stage addressed the problem of how to improve the existing fleet on the West Coast, while the second stage assumed that a new vessel could be designed and constructed to maximize the fisherman's profits by increasing the ratio of tonnes of fish caught per dollar expended. In 1983, Dr. S.M. Calisal and Otto Fung [1] produced a computer simu-lation for fishing vessel operations which would accurately predict the fuel costs for displacement type vessels operating on the West Coast. The computer model included methods to decrease fuel costs, that could be retrofitted to the existing ships. The methods included, controllable pitch propellers, two speed gearboxes, and Kort Nozzles, which would be added to the vessel and possible fuel savings, for a given fishing 2. scenario, for each method were calculated. This program can effectively make recommendations to the ship owners, about which method of fuel saving is best to his vessel, but i t Is f e l t that the operators should decide on the method that is most suitable for their operations. The study to design an optimum fishing vessel for a specific operating scenario was seen as the solution to the West's problem of how to replace the aging vessels on the Pacific Coast. This optimization study differs from past studies, in that i t focusses specifically on the small trawler (under 39 m in length) which is indigenous to the West Coast of Canada, while past studies have been centered around cargo ships or tankers. In 1963, Murphy, Sabat and Taylor [2] developed a systematic procedure to determine the optimum combination of design variables for a cargo ship that would satisfy the owner's requirements using the least cost criterion. Paterson [3] in 1984 used this least cost criterion to determine the best design variables for fishing vessels on the Pacific Coast. Mandel and Leopold [4] in 1966 used the least cost criterion, and adopted i t to a computerized optimization technique that would determine the best vessel dimensions, based on the owner's requirements and the operating scenario for a cargo ship. The f i r s t optimization study applied to fishing vessels was done by Kupras In 1966 [5]. His model was based on Polish Factor Tankers and used methods similar to those previously mentioned. In 1971, Hamelin [6] used net parameters, vessel parameters, and least time in port as the criterion to determine the best vessel to be used for ground fishing on the Atlantic Coast. In the following chapters, the author presents a method to determine optimum dimensions of a trawler, based well as the lowest operating costs and intention of this thesis i s to produce be used by fishing vessel designers. 3. on a fixed operational scenario as largest fish hold capacity. The a preliminary design tool which can 4. CHAPTER 2 FISHING VESSEL DESIGN MODEL The design principles applied to fishing vessels are the same as those used to design any ship. The building and annual operating costs can be f a i r l y accurately predicted i f the following design variables are known: 1. Ship Length, L 2. Ship Beam, B 3. Ship Draft, T 4. Ship Depth, D 5. Midship Coefficient, CM 6. Prismatic Coefficient , CP 7. â€¢ Ship Displacement, A or Volume V 8. Required Horsepower, SHP 9. Ship Speed, VS This l i s t of nine variables can be reduced due to the interdependence of some of the variables. Wilson [9] has shown that, for fishing vessels, freeboard is a function of length, since depth is the summation of draft and freeboard, i t can be eliminated i f the draft of the ship i s known. Displacement or volume are dependent on length, beam, draft, midship coefficient and prismatic coefficient and therefore can be eliminated i f these five quantities are known. The required horsepower of a ship can be expressed as a function of the ship dimensions and speed and is therefore eliminated i f ship speed is known. By using these assumptions, the original l i s t of variables i s reduced to the six variables which are defined as the ship vector: SHIP = f ( X l , X2, X3, X4, X5, X6) 5. where XI = ship length X2 = ship beam X3 = ship draft X4 = midship coefficient X5 = prismatic coefficient X6 = speed-length ratio If the owner of the vessel were to specify a required speed, the variable l i s t would be further reduced to five. Since i t i s assumed that there is a vessel that w i l l efficiently complete the operational scenario, the ship speed i s considered as one of the variables, which implies that the operational scenario is speed dependent. The midship coefficient and prismatic coefficient were included in the ship vector in order to define the volume and displacement of the vessel as well as being required input for the calculation of the ship's resistance. In order to ascertain whether or not the optimum ship is capable of safely conducting fishing, the design procedures must include some constraints. This study uses vessel equilibrium, stability and volume available for catch as the design requirements. Vessel equilibrium is defined as the difference between the ship displacement, (calculated from the ship vector) and the summation of the weights of equipment and f a c i l i t i e s required to conduct fishing operations. In equation form: Equilibrium = A - E where A = W i " 6 . ship displacement individual weight groups (e.g. machinery, fuel, etc.) The displacement must always be greater than the sum of the weights or the vessel w i l l sink, therefore this requirement must be satisfied. The s t a b i l i t y of the ship is included to ensure that the optimum vessel does not converge to a point where there i s zero beam value, which could happen i f optimum vessel were chosen based on minimum fuel cost. This requirement also indicates that the ship w i l l be capable of working at sea safely. Volume available for catch, also termed Fish Hold Capacity (FHC), i s the criterion that determines whether or not the ship can earn money. Fish hold capacity i s defined as the difference between the vessel volume and the summation of the volume required for equipment and f a c i l i t i e s . In equation form: FHC = V - Z VOLi where V = vessel volume VOLi = volume groups (e.g. machinery, crew, etc.) 2.1 Relationship Between the Ship Vector and Ship Dimensions The following eleven equations show the relationship between the design variables and the ship dimensions. 1. XI = Ship length (m) = L 2. X2 = Ship beam (m) = B 7. 3. X3 = Ship draft (m) = T AM = Midship area (m2) = X4 * X2 * X3 5. VOL = Ship volume (m3) = X5 * XI * AM 6. DISP = Ship Displacement (tonne) = 1.0252 * Vol 7. Vs = Ship Speed (knots) = X6 */Xl * 3.2808 8. FBD = Ship Freeboard [9] (m) = Xl/35.2 * .270 XI x 3.2808 is conversion from metres to feet 2.2 Limits Assigned to Components of Ship Vector The range of values shown below are the actual limits of the variables used for vessels on the Pacific Coast. They are based on information obtained from local Naval Architects [10] and their designs over the past 20 years. 16.0 m < XI < 35.5 m 7.11 m < X2 < 11.2 m 1.92 m < X3 < 3.95 m .411 m < X4 < .915 .495 m < X5 < .695 .617 m < X6 < 1.31 9. D = Ship depth at midships (m) = X3 + FBD 10. CWP = Waterplane Coefficient [9] - .65 * X5 + .395 11. AW = Waterplane area (mz) = CWP * XI * X2 2.3 Equations Used for Fishing Vessel Design Model 8. Prior to attempting to find the optimum fishing vessel one must be confident that the approximations used in the design process are accurate. The design procedure used in this thesis requires the calculation of the ship's resistance and power. Since the requirements for equilibrium, sta b i l i t y and fish hold capacity exist, the accurate determination of the weight groups, volume required by each group and the s t a b i l i t y of the vessel must be made. The optimization criterion (explained later) requires that the construction costs and annual operating costs be calculated in order to determine the ratio of the costs to the fish hold capacity. The equations used in the fishing vessel design model are presented in five (5) sections: I) resistance and power, i i ) weights, i i i ) volumes, iv) s t a b i l i t y , v) costs. 2.3.1 Resistance and Power The estimation of a ship's resistance is a computerized version of Oortmerssens' Method [11] for predicting small ship resistance, which i s based on 930 resistance data points taken from 93 models of tugs and trawlers tested at the Nederlands Ship Model Basin. This algorithm was chosen over other available algorithms because i t required the minimum number of Input variables (compared to the other algorithms) to accurately predict to total resistance of the ship [12]. The resistance algorithm is shown below: RT = RR + RF where 9. RR - [C1 e ^ , g + e"â„¢ [C 2 + Cg sin(FN" 2) + cos(FN" 2)]] * DISP RF = [.075/(43429 An RN-2)2 + .00051] + R0/2 * V 2 * S and CI = di,0 + di.l(LCB) + di,2(LCB) 2 + di,3(CP) + dl,4(CP) 2 + dl,5(L/B) + di,6(L/B) 2 + dl,7(CWL) + di,8(CWL)2 + di,9(B/T) + di,10(B/T) 2 + di,ll(CM) S = 3.223 * V0L 2 / 3 + .5402 * L * V0L 1 / 3 dl,12(i = 1,4) are shown In Table I. .00051 = Hull Roughness Correction Factor m = .14347CP" 2* 1 9 7 6 + FN"2 Table I Values of d i , l l ( i = 1,4) i = 1 2 3 4 di.O 79.32134 6714.88397 -908.44371 3012.14549 d i , l .09297 19.83 2.52704 2.71437 di,2 .00209 2.66997 .35794 .25521 di,3 -246.45896 -19662.024 755.1866 -9198.8084 di,4 187.13664 14099.904 - 48.93952 6886.60416 di,5 - 1.42893 137.33613 - 9.86873 - 159.92694 di,6 .11898 13.36938 .77652 16.23621 di,7 .15727 4.49852 3.7902 .82014 di,8 .00064 .02100 .01879 .00225 di,9 - 2.52862 216.44923 - 9.24399 236.3797 di,10 .50619 35.07602 1.28571 - 44.1782 d i . l l 1.62851 - 128.72535 250.6491 207.2558 10. 2.3.1.1 Effective Horsepower [14] Effective horsepower is the power required to move the bare hull through s t i l l water at a given design speed RT * V EHP = 550 where RT = total ship resistance in lbs. V = ship speed in ft/sec. 550 = conversion from ft-lb/sec to HP 2.3.1.2 Required Shaft Horsepower The required horsepower i s the power required to propel the ship at the desired speed and includes propeller and shafting bearing losses, as well as a correction for sea conditions. Refrigeration requirement is added as a constant to the required horsepower [7]. S H P = .97 * OPC + * 3 E H P + 4 5 where .97 = estimated transmission efficiency OPC = overall propulsive coefficient â€¢3EHP = correction for sea conditions [13] 45 = refrigeration constant [7] 2.3.1.3 Overall Propulsive Coefficient [14] The efficiency of the ship propulsion system i s given by the overall propulsive coefficient (OPC) where EO = KT * J/KQ x 2 ir ER - 1.02 and Propeller performance coefficients KT, KQ and J are calculated during the propeller selection. 2.3.2 Weight Equations The accurate estimation of weight of each group i s required in order to effectively determine the vessel equilibrium. The cost estimation i s heavily dependent on the steel weight used in ship, therefore the estima-tion of steel weight is very c r i t i c a l to the correct analysis of the optimum vessel. The estimation of weight required for the fishing vessel i s divided into the following individual weight groups: Steel Weight: Includes; hull, superstructure and appendages (excluding propeller shaft and rudder stock). Outfit Weight: Includes; auxiliary machinery, piping with liquid e l e c t r i c a l , joiner work, furniture, hull outfit (i.e. firefighting and rescue equipment), fishing outfit, spars, rigging and paint. Machinery Weight: Includes; main propulsion system, graving, controls, propeller and shafting, pumps and main switchboard. Fuel Weight: Includes; fuel o i l and lubricating o i l required t complete nine days at the design speed. 12. Fresh Water and Provisions: includes; fresh water and provisions required for the nine day t r i p . 2.3.2.1 Estimation of Steel Weight (WS) Accurate estimation of steel weight i s d i f f i c u l t to calculate for West Coast trawlers, as there is not much information available about the invoiced steel weight used i n the ships. The method used i n this thesis i s based on the invoiced steel weights used in three classes of vessels constructed i n Vancouver, as well as three formulations for weight estima-tion, for small ships, from eastern United States and eastern Canada. The steel weight i s calculated using each of the formulations shown below, then the average of the methods depending on length is taken to give the steel weight estimate. The decision to use the average of the estimates was a consequence of each method having weaknesses at either end of the range of lengths. Appendix A details the percent error i n each method relative to the invoiced steel weight used for the three classes of vessel. Three methods used i n the formulation of the steel weight estimate are shown below: a) East Coast Formulation [15] WS1 2.813xl0~5 (X 3 - 173.52X2 + A.lSSxIO'+X - 1.58xl0~5) where X = UB+D) 1 u / q 3000 1 * 100 and L,B,D are in feet WS1 in long tonnes. b) Santarelli Formulation [13] 13. WS2 ( 3 ) = .037 E 1 ' 3 6 where E = L(B+T) + ,85L(FBD) + â€¢85Zl 1h 1 '+ .75 Z l 2 h 2 and .SSEljhj + .75Il 2h 2 are superstructure terms For small ships the superstructure terms can vary from 0-150. In this case for L < 29.5 m superstructure = 50; for L > 29.5 superstructure = 100, which implies that: WS2 = .037(L(B+T) + .85L(FBD) + 50) 1* 3 6, and, WS3 = .037(L(B+T) + .85L(FBD) + 1 0 0 ) 1 , 3 6 c) Wilson Formulation [9] WS4 = CSTRUCT x CUBIC NO. x 1.0163 where CUBIC NO. CSTRUCT The decision to use the average of the formulation was made by calcu-lating the estimated steel weight for each formulation for vessel lengths between 16 m and 40 m and plotting steel weight vs. length (Fig. 4). Since each formulation uses length, beam, and depth to calculate the weight, the following parametric equations, which were derived from plots of beam vs. depth and beam vs. length for West Coast vessels (Figs. 5 and 6), were used to determine the values of beam and depth corresponding to each length. LxBxD + .326 2.834 CUBIC NO. 50,000 14. i) B = i i ) L = Using this method the steel -.0642D2 + 2.1019D + .3683 .68553B2 - 3.8933B + 16.123 weight estimate i s : For L < 29.5 m WS1 + WS2 + WS3 WS = 3 and for L > 29.5 m WS2 + WS3 + WS4 WS = 3 2.3.2.2 Outfit Weight (WO) The calculation of outfit weight i s based on the Wilson Formulation [9], which was derived from plotting the weights of the items which make up the outfit group, as specified previously. WO = COUT * CUBIC NO. * 1.0163 LxBxD where CUBIC NO. = 2 â€žâ€ž T t ( r CUBIC NO. . COUT = - . n + .196 17.140 2.3.2.3 Machinery Weight (WM) The weight of the main engine and gearing i s known from the manufac-turers data sheets, but the associated machinery (pumps, etc.) is not known. Therefore, i t i s necessary to estimate the machinery and equipment, Santanelli [13] has derived a relationship between weight and engine para-meters, from known weights for vessels constructed in Spain. This formula-tion seems to give f a i r l y accurate results when compared to the known engine and gearbox weight. This method gives a machinery weight of 3 to 6 tonnes higher than the catalogue weights, which would be sufficient to include the pumps and other machinery used in a ship. Other formulations 15. that were available usually gave weights equal to or lower than the catalogue weight WM = CMY (â€¢ MCR RRPM .75 [13] where MCR rated horsepower RRPM rated RPM CMY 20 for MCR < 1000 MP 30 for MCR > 1000 MP 2.3.2.4 Fuel Weight (WF) The fuel weight i s calculated from the actual fuel consumption curves for the selected engine. Fuel required (FR) i s based on nine days at continuous design speed. The assumption implies that the fuel weight estimate w i l l be higher than the fuel used, because the speed of the vessel i s usually reduced on the homeward leg, but i t is f e l t that i t i s required to allow extra fuel for fishing operations in varying sea conditions. FR = 9 days * FCR (gal/hr) where FCR = fuel consumption ratio (gal/hr.) .*. FR = 9 days x 24 hrs/day * FCR * .0045 m3/gal It is assumed that the specific weight of fuel is .85 x fresh water weight therefore: Wt of Fuel = .85 * FR (tonne) The weight of lubricating o i l i s assumed to be 1 percent of fuel required therefore: WF = 1.01 * Wt of Fuel (tonne) 2.3.2.5 Weight of Fresh Water and Provisions (WWS) The weight of fresh water i s derived from Santarelli's fresh water consumption curve [13], based on crew size, length of vessel, and daily consumption. Fresh Water Consumption (CFW) CFW = [.7667L + 6] * N * Days * 10~3 m3/Jlt Wt of fresh water = 1 t ? n n e * CFW mÂ° Weight of provisions is assumed to be 5 times the weight of fresh water, which w i l l allow for meats, frozen provisions, and canned goods. The total weight of fresh water and provisions i s : WWS =Â» 6 * CFW The total required weight of the fishing vessel i s : WTOT =WS+W0+WM+WF+ WWS and must be less than the calculated displacement of the vessel (up to the design waterline) to ensure that the ship w i l l f l oat. 2.3.3 Volume Calculations The estimates for the volume required are derived in terms of percentage of waterplane area dedicated to each compartment. The decision to use percentage of waterplane area to approximate the volume was based on the fact that space requirements (i.e. crew space, machinery space, etc.) are specified in terms of square footage. Since waterplane area Is a function of length and beam, i t was decided to use the ratio of compartment length to waterline length; to determine the percentage of area dedicated to each compartment. Fig. 1. Simplified Waterplane Arrangement for Fishing Vessels. LB P Note: Ratio of length of compartment to waterline length is for example [XER/LEP]. 18. This ratio i s based on measurements taken from twenty-five general arrangement drawings for vessels constructed on the West Coast during the last twenty years [3] [16]. The ratio of lengths i s multiplied by the waterplane area and the draft to approximate the volume required for each compartment. This method for approximating volume i s satisfactory as long as i t is assumed that the changes in prismatic coefficient are small, to ensure that total volume changes are small. The required volume of the vessel is the sum of the five groups: (1) Fuel volume; (2) Machinery volume; (3) Fresh water volume; (4) Stores volume; (5) Crew living volume; and must be less than the vessel volume which i s defined by: VESSEL VOL = [CP x CM x L x B x T] + [AW + FBB] in order to have space available for fish. The design requirement for fi s h hold capacity is the motivation for including the volume calculations in this study. Fuel Volume The volume of fule required for the nine day t r i p i s based on the fuel consumption ratio of the engine: FUVOL = FR Machinery Volume (MACV) The average ratio of lengths of machinery compartment to waterline length i s : .2344 based on 24 ships that have been designed and constructed 19. on the West Coast [10][16]. This ratio i s multiplied by the waterplane area and draft to approximate the volume required for machinery. It is common in fishing vessels to have the fuel tanks integral with the engine rooms therefore machinery volume is defined by: MACV = .2344 * (AW * T) - FUVOL Fresh Water Volume (FWVOL) The volume of fresh water required for the scenario i s the volume of water consumed per man per day. Stores Volume (STVOL) The estimation of the volume required for stores includes the space required for the galley. This formulation is based on the values given by Santarelli [13] for stores requirement in a fishing vessel. STVOL = 16.96 x N Crew Living Volume (CLVOL) The volume required for the crew i s based on the livi n g space requirements stated by Santarelli [13], and the ratio of lengths measured. It must be noted that only a portion of the livi n g space i s below the waterline, and most of the living space is housed in the superstructure of the vessel. 20. Fig. 2. Simplified Side View of Fishing Vessel Showing Idea of Accommodation Below Waterline. Note: This is f a i r l y typical of West Coast vessels. CLVOL = .2155 * CLAREA ^ VOL Waterplane Area where CLAREA = crew living area [12] = 30 m2 for N=l = 30 + N * 20 for N>1 .2155 is ratio of crew space to AW below DWL. The total required volume for the vessel i s : TVOL = FUVOL + MACV + FWVOL + STVOL + CLVOL 2.3.4 Stability Since the approximations for the centre of gravity for each weight group are very inaccurate, and the actual values were unavailable, i t was decided to use the Wilson Formulation [9] to determine the vessel's metacentric height. Wilson derived a curve from a plot of GM/B versus length which gives metacentric height as: GM = B * (-L/400 + .185) This formulation was also used by Latore [17], when he compared three s t a b i l i t y regulations that are applicable to fishing vessels. In order to ensure that the vessel w i l l survive at sea, the design must satisfy both; the International Consultative Organization (IMO) requirements [18], as well as the Japanese fishing vessel requirement [19]. The decision to satisfy both requirements was based on Latore's [17] findings which Indicated that each method had i t s shortcomings. The required metacentric height for the vessel i s : 1. IMCO Requirement GMR = 1.7388 + 2.0 x B x (GM1 + GM2)/3.2808 where: GM1 = .075 - .37(FBD/B) + .82 x (FBD/B)2 GM2 = -.014 x (B/T x FBD) and 2. Japanese Requirement [19] GMR is the greater of: GM1 = (B - 7.0/12.0) + .4 GM2 = (1 - 4.2/72.0) + .4 22. 2.3.5. Cost Calculations Accurate estimation of the costs involved in the construction and operation of a fishing vessel are required to be able to determine with confidence that the recommended vessel i s truly a superior ship. This is d i f f i c u l t to accomplish because most shipyards are unwilling to supply the actual cost breakdown for vessels constructed in their f a c i l i t y . Mr. Aurthur MacLaren, President of Allied Shipyard in North Vancouver [23], was very helpful without being too specific, by detailing the construction costs (in his yard) in terms of the "rough" percentages of each group to the total vessel cost. The following assumptions were made based on the "rough" estimate. The cost of materials (in this case steel) tried to construct the vessel can be taken as approximately 10 percent of the total cost. The labour cost to the erect the steel is seen to be approximately 20 percent of the total cost. While the cost of outfit and labour for outfit make up the remaining 70 percent. It must be understood that in this context "outfit" means a l l equipment that i s purchased outside the yard, for example, engines, radar, etc. Every fishing vessel owner has his own preference for type of engines, radars, radios and other equipment, so the test of accurately predicting the cost of outfit equipment i s very d i f f i c u l t . The cost of steel, on the other hand, is readily available, and therefore by using the assumption that the steel used in the vessel accounts for 10 percent of the total cost, an accurate prediction of the construction cost can be made. This assumption also makes i t imperative that the approximation for steel weight is as accurate as possible. By calculating steel weight for the vessel and using existing steel costs the following relationship i s derived: 23. STEEL COST = 638.19 x WS where 638.19 = conversion from 508.00 per 2000 lb tonne to metric tonne [24] WS = Steel weight Since i t is assumed that materials cost i s 10% of the vessel cost the construction cost for completed vessel i s taken to be: CC0ST = 6381.9 x WS This estimate of total vessel cost appears to be f a i r l y accurate when compared to the rough "benchmark" of $1000.00 per foot of length current used in the industry [23]. For example, a class of 76 f t . vessels constructed in Vancouver averaged approximately $750,000 (f i n a l price) and the predicted cost using the method prescribed above came to $696,000 based on steel weight. The construction cost i s used to determine the fixed annual operating costs which Include: mortgage, depreciation, insurance, repairs, licence, port fees, and wages. The total annual costs are divided into the operational (fixed) costs and the operating (variable) costs, which include fuel and stores. The total cost i s the adjusted on a per t r i p basis, assuming that the vessel spends 225 days at sea, or 25-nine day fishing trips. Operational Costs (Fixed) The fixed costs are termed operational for the simple reason that in 24. order to proceed to sea to conduct fishing operations these costs must be covered. These costs are based on present mortgage rates, depreciation rates, and insurance rates obtained from banks, surveyors and insurance companies in the Vancouver area. It is also assumed that the ship w i l l have a twenty year l i f e - c y c l e . Annual Mortgage Calculation This calculation i s based on 25 percent down payment (i.e. 25 percent of construction cost) and an annual rate of Prime + 2 percent [20] AMR = annual mortgage rate = Prime + 2 percent PM = monthly payment (1+X)Y Y . = â€” -zâ€” x X x A (1+X) -1 where X - AMR/12 Y = 240 (12 x 20 payments for vessel l i f e ) A = .75 x construction cost. Annual Mortgage Cost (CMO) = 12 x PM Depreciation Calculation This calculation i s based on 5 percent straight line depreciation with 10 percent salvage value after 20 years [21]. Depreciation Cost (DC) = .045 * Construction Cost 2 5 . Insurance Rate The Insurance rate i s dependent on the total cost of the vessel [22]. The Insurance costs as a function of the i n i t i a l vessel cost are given below: Cost of Vessel Under $500,000 CI = .02 x Construction Cost Cost of Vessel Over $2,000,000 CI = .01 x Construction Cost $500K < Cost of Vessel < $2M CI = .015 x Construction Cost Miscellaneous Cost These costs include repairs, port fees, licence, and wages, and are dependent on how successful the year had been. If a vessel had an unsuccessful year, very l i t t l e i s spent on repairs for example, so i t is assumed that the miscellaneous cost would be: CM = .2 x [CM0 + CD + CI] The total annual operational costs are therefore: Operational Cost = 1.2 x [CM0 + CD + CI] Operating Costs (Variable) The operating costs are the actual cost incurred while fishing, these are basically fuel cost and provisions. The fuel cost is the actual cost of fuel used during the nine day scenario calculated at a fuel cost rate of 38.5 cents per l i t r e . The cost of provisions is assumed to be $1500.00 per tonne, based on $1.5/kg average price for groceries and cleaning gear. The provision cost is rated on a consumption basis. 26. CHAPTER 3 THE OPERATIONAL SCENARIO A fishing vessel, unlike a cargo ship, has a complicated set of operating conditions. The operational scenario of a cargo ship i s basic-all y , load cargo in harbor, start engines, s a i l at constant speed to the delivery port, then stop engines. On the other hand a fishing vessel considered here, usually has a high speed, light load condition during the transit to the fishing ground, a changing displacement, low speed high drag fishing operation, then a high speed deep load condition to return to the packing plant. Figure 1 shows a possible operational scenario for a trawler, that leaves a packing plant, conducts fishing operations, and returns to the same packing plant. The duration i s determined by the fact that the fisherman must return to the plant within seven days of his f i r s t catch, or the f i r s t day catch Is spoiled and worthless [7]. The typical fishing trip is shown in three phases. During the transit to the fishing ground (phase 1) the vessel i s assumed to be at design conditions and travels at the design speed for two days. The fishing operation (phase 2) are simulated by decreasing speed to 3 knots (standard travel speed) and adding net resistance to the vessel resistance at 3 knots, for 4 days. On the return t r i p (phase 3) the vessel displacement i s increased and speed reduced from the design condition un t i l the total resistance of the vessel on return i s equal to the design resistance. If the speed is reduced to such a point that the return time is greater than 3 days the vessel speed i s increased un t i l the ship travels back to the packing plant within 72 hours. During each phase the fuel consumption i s 2 7 . Fig. 3. Operational Scenario Â© TIME ( Days ) Â© 8or9 calculated, then added together to arrive at the total fuel consumption for the t r i p . The calculation of fuel consumption i s based on the actual fuel consumption as engine RPM curves found in the Caterpillar engine manuals. 28. CHAPTER 4 OPTIMIZATION CRITERIA It is assumed that a fishing vessel can be designed and built to economically conduct the operations dictated by the scenario. In order to determine which vessel is best for the scenario, the optimization c r i t e r i a must be selected as the best measure of the vessels earning capability. The c r i t e r i a for optimization is also dependent on the design requirements. If, for example, the design requirement stated that the fish hold capacity would be 'X m3', then the optimization would be based on a least cost c r i t e r i a , meaning that the best vessel would be the vessel that expended the minimum amount of money for the specified fish hold. The design process in this thesis i s based on the operational scenario, which means that the optimization c r i t e r i a must include, in some form, the income and expenses associated with the vessel. The earning capability of the vessel i s derived from the ship design model. For any given set of design variables, a specific set of ship dimensions are calculated, which define the volume and displacement of that vessel. The sum of the weight groups in subtracted from the displacement of the vessel and i s a measure of the ship's a b i l i t y to float, that i s i f the difference between displacement and the weight required i s positive the ship w i l l float. The difference between the sum of the required volumes and the vessel volume i s in fact the volume available for f i s h holds. This avail-able volume i s termed the fish hold capacity (FHC) and is the measure of the vessel's earning capability. The total annual operating costs for the given set of design variables are also calculated, which are the measure of the expenses associated with 29. that ship. Since the optimization c r i t e r i a i s required to measure the vessel's earning capability relative to the expenses, i t was decided that the cost merit function (CMF) or optimization c r i t e r i a should be the total annual operating costs divided by the fish hold capacity. This means that the cost function would have the units of dollars per cubic metre, but more Importantly, by defining the cost function in this way, the optimization c r i t e r i a i s independent of the type of fis h caught, or the amount of fish caught per t r i p . This Implies that the best fishing vessel for the scenario would be the ship that has the lowest cost function, which i s the vessel with the lowest operating costs and largest fish hold. The merit function i s therefore dependent on the design process and the operational scenario and is defined by: CMF = f(Ship Vector, Scenario) which is calculated for each set of design variables. 4.1 Steps Involved in Evaluating the Cost Function The following steps are required to evaluate the cost function: Step 1: Select i n i t i a l t r i a l values of the components that make up the ship vector. Step 2: Calculate the ship's dimensions and coefficients, estimate resistance, and calculate the effective horsepower. Step 3: Select the propeller, based on Wagengen B series, calculate overall propulsive coefficient, and calculate the required horsepower [14]. 30. Step 4: Select main engine based on required horsepower, and select gearbox to match propeller RPM and engine RPM. The engine and gearbox selections are based on Caterpillar marine engines [8]. Step 5: Calculate the required weights and volumes, and check that the sum of weight groups and the sum of volume groups do not exceed the displacement and volume of the vessel. Step 6: Calculate the i n i t i a l s t a b i l i t y of the vessel and check that i t satisfies the stability requirements. Step 7: Simulate fishing operations and calculate the fuel required for each phase of the t r i p . Step 8: Calculate the construction, operational and operating costs for the t r i p . Step 9: Evaluate the cost function and output vessel design variables. The equations used at each step are defined in the Fishing Vessel Design Model section. ! 31. CHAPTER 5 VALIDATION OF DESIGN PROGRAM The design program was tested by using the actual variables associated with "Eastward Ho", a trawler owned by Mr. Gary Sigmund [7]. This vessel was also used by Calisal and Fung to calibrate the fishing vessel cost program. In the case of the cost program, the actual vessel parameters are inputs, that is the ship dimensions, engine type, speed, propeller type, and net resistance. The owner then conducts a simulated fishing t r i p by moving his ship (simulated by moving the cursor on the computer terminal) to the fishing ground from his home port, then returns to his home port. After comparing the actual fuel cost incurred by the owner, to the cost predicted by the program, there was less than 4 percent difference. In this program the design varibles (Xi i-l,6) "Eastward Ho" were used as inputs, and the output results, shown in Table II, were compared to "Eastward Ho". Table II Comparison Between Eastward Ho and Values Predicted in the Design Program Ship Pariculars Eastward Ho Design Program Length 29.26 m 29.26 m Beam 8.894 m 8.894 m Draft 2.926 m 2.926 CM .5251 .5251 CP .8241 .8241 V//T7 1.0206 1.0206 RT 10,970 lbs. 10,366 lbs. Engine Type D398 CAT D398 CAT Rated Horsepower 850 HP 850 HP Gear Ratio 3.92 4.00 Propeller (Bar) 4 Blade (.55) 4 Blade (.55) Diameter 6.0 f t 6.23 f t Fish Hold Capacity 10,430 f t 3 9,065 f t 3 32. The validation was done i n order to determine the accuracy of the weight and volume approximations. The program is f a i r l y accurate at predicting the values of fis h hold capacity, as there i s only 13 percent difference between the actual fish hold capacity in "Eastward Ho" and the fi s h hold capacity estimated by the design program. The program selects the same engine that is in the ship, as well as the same transmission. The gear ratio i s different because the new transmission model does not include a 3.92:1 gear ratio. It is evident from the results that the program can be used at the preliminary design stage to f a i r l y accurately design a fishing vessel. 3 3 . CHAPTER 6 SENSITIVITY ANALYSIS To determine what effect changes in the design variables would have on the merit function, each variable was individually changed by Â±15 percent (from the i n i t i a l values of "Eastward Ho") while the remaining five variables were held constant. From the following equation: 3A 8L 8B 9T ... â€” = r + r + â€” [14] i t is evident that any change in the major dimensions of the vessel w i l l directly affect the vessel volume. For changes in length, i t i s assumed that the change in fish hold capacity is equal to the change in vessel volume, as changes in length are viewed as additions or subtractions of middle body length which are where the fish holds are usually located. Changes in beam and draft have the same effect on vessel volume that changes in length have, but the change in fish hold capacity due to changes in beam or draft can only be affected i n the refrigerated compartment. This means that for a 10 percent increase in either beam or draft the vessel volume i s increased by the same amount as a 10 percent increase in length, but the fish hold capacity increases less than 10 percent because the refrigerated space does not occupy the entire volume of the vessel. The percentage change of fish hold for beam and draft changes is derived from measurement of f i s h hold lengths (in the 25 general arrangement draw-ings) and dividing by the waterline length. Over the past 25 years the fi s h hold/waterline length ratio i s .4283. This ratio i s then multiplied by the change in beam or draft relative to the original beam or draft to 34. obtain the percentage change of f i s h hold capacity for changes in beam or draft. Change in volume due to changes in midship coefficient and prismatic coefficient are based on the same assumptions as f i s h hold capacity changes due to changes in beam or draft. The effects that changes in the design variables have on the cost function are shown i n Figs. 7 through 11. The effect of changes in the design variables on fuel cost are shown i n Figs. 12 through 16. 6.1 Cost Function vs. Changes in Length The cost function increases with increases in length as shown in Fig. 7 which is the curve for a specific vessel (Eastward Ho). There are local minimums at 27 m and 30 m, caused by crew size increases. Since crew members are integer values, the volume required for one additional crew member is greater than the vessel volume increases for the change in length, hence the fish hold capacity decreases slightly to cause a lump i n the cost function curve. The trend of increasing cost function with length increases is caused by the vessel costs rising ( i . e . construction and fuel) faster than the f i s h hold capacity increases therefore causing the cost function to rise. 6.2 Cost Function vs. Changes in Beam A narrow beam vessel implies low resistance, which means decreased i fuel costs, but i t also means that the fish hold capacity is small, hence a high cost function. As beam increases the f i s h hold capacity increases, without affecting the total cost that much therefore the cost function decreases, u n t i l the beam gets large enough to affect the power 35. requirement. As shown i n Fig. 8, a beam greater than 9.7 m requires extra power, which increases fuel costs greater than the fish hold capacity increases therefore the cost function climbs to a higher level. 6.3 Cost Function vs. Changes in Draft The cost function i s 96 dollars/m 3 at a draft of approximately 2.3 m, because of minimum fis h hold capacity. Fig. 9 shows that the cost function decreases until the power required increases (because of Increased draft) which causes the total cost to rise quicker than the fis h hold capacity increases therefore causing the function to increase. 6.A Cost Function vs. Changes in Midship Coefficient Since the midship coefficient Is a measure of the fullness of the middle body, the same trend in the cost function that i s evident with changes in beam and draft is expected. Fig. 10 shows that the cost function i s high at low midship coefficient values, (because of small f i s h hold capacity) and decreases to a point where the cost to move the vessel i s greater than the Increase In fi s h hold capacity causing i t to rise again. 6.5 Cost Function vs. Prismatic Coefficient When the prismatic coefficient is varied, the cost function shows a similar trend to that of the change in length curve. This i s explained by the fact that the prismatic coefficient is a measure of the volume of the 36. vessel and as such i s dependent on the major dimensions of the vessel. It is evident in Fig. 11 that changes in the prismatic coefficient affect the cost function least of a l l the variables because the range of cost function values (Y axis) is small. 6.6 Fuel Cost vs. Changes in Design Variables One would expect that i f the design variables are changed (increased) that fuel costs would increase, due to increase resistance of the vessel. Figs. 12 through 16 show that as each variable is changed the fuel costs increase accordingly. 37. CHAPTER 7 ROUTINE FOR OPTIMIZATION The optimization routine used in this thesis is the Coupler Optimiza-tion Technique, which originated i n the UBC NLP Library, was modified to operate on the department's VAX 11/750 system. This routine was chosen because i t was the most flexible optimization routine i n terms of inputs required. It is necessary to provide only the optimization function, (which i s the design program) without partial derivatives, as well as the upper and lower limits assigned to each variable in the optimization function. It also guarantees fast, accurate convergence for optimization functions with less than 10 variables. Since the optimization function is the design program, which calculates the cost function for each vessel, the optimum vessel for the fixed scenario would be the vessel with the lowest cost function. In order to determine whether or not the vessel was satisfactory the following conditions are required to be met before the cost function i s calculated: i) Ship displacement must be greater than the sum of the weight groups; i i ) Vessel volume must be greater than the sum of the volume groups; i i i ) The stability requirements must be satisfied; iv) The ship must be capable of pulling a net, and must be capable of returning to port with holds 65 percent f u l l . If any one of these conditions are not met, the program sets the cost func-tion at a very high value, which implies that the vessel i f not acceptable. The complete optimization, when run on the department VAX system, takes 23 iteractions and between 7 and 8 minutes of real time (1-1.5 min CPU time). 38. 7.1 Results of the Optimization The optimization program was run using the values for "Eastward Ho" as the i n i t i a l values of the ship vector. The program predicted that the following combination of design variables would decrease the cost function by 22.5 percent: XI = 30.395 m X2 = 9.704 m X3 = 2.948 m X4 = .874 X5 - .544 X6 = .964 To ensure that the design variables were a true optimum the optimized values were used as the new input variables for the second run on the optimization. The program gave the identical results for the second run. The program was also run with random input variables, ( i . e . anywhere within the limits) and the same values were returned as the optimum. The program was also run for a 16 m steel trawler, and the predicted values had the same trend as those of the "Optimum" "Eastward Ho". That is the cost function decreased with slight increases in the design variables. Table III shows the changes in design variables after optimization for the 16 m trawler. These two vessels were the only ships tested, as they were the only vessels that had complete information available. 39. Table III Results of Optimization for 16.00 m Vessels Design Variables Original Optimum XI 16.00 m 18.39 m X2 4.92 m 5.17 m X3 1.68 m 1.80 m X4 .81 .836 X5 .64 .655 X6 1.38 1.20 7.2 Comparison of Optimized Ship to Eastward Ho To compare the optimum design values to those of "Eastward Ho" the optimum values were used as inputs In the design program. Table IV details the parameters of both ships. Table IV PARAMETERS EASTWARD HO OPTIMUM VESSEL Design Variables XI 29.26 m 30.395 m X2 8.894 m 9.704 m X3 2.926 m 2.948 m X4 .824 .874 X5 .525 .544 X6 1.021 .964 Ship Dimensions MIDSHIP AREA 21.45 m2 25.00 m2 VOLUME 329.51 m3 413.52 m3 DISPLACEMENT 337.81 tonne 423.99 tonne SPEED 10.00 kts 9.63 kts Waterplane Area 191.62 m2 220.84 m2 4 0 . Table IV Cont'd PARAMETERS EASTWARD HO OPTIMUM VESSEL Ship Resistance RR 8340.4 lbs. 8030.7 lbs. RF 2025. lbs. 2152.9 lbs. RT 10,366.02 lbs. 10,183.56 lbs. EHP 318.32 HP 301.20 HP Propeller Parameters DIA 6.24 f t . 6.29 f t . RPM 276.31 268.33 EFF .5536 .5476 THRUST 11,746.88 lbs. 11,582.57 lbs. Overall Propulsor Coeff. .6212 .6144 Required Horsepower SHP 668.80 HP 640.73 Engine Parameters TYPE CAT D398 CAT D398 RRPM 1225 1225 RHP 850 HP 850 HP BHP 705 HP 705 HP FCR 40.06 gal/hr. 40.06 gal/hr. EGR 4.00 4.00 Weight Estimates WS 134.98 tonne 150.74 tonne WO 80.46 tonne 93.79 tonne WM 15.21 tonne 15.21 tonne WF 33.78 tonne 33.78 tonne WWS 7.68 tonne 7.91 tonne Volume Estimates MACV 92.13 m3 113.33 m3 FUVOL 39.34 m^ 39.34 FWVOL 1.28 m3 1.32 m3 STVOL 84.80 m3 84.80 m3 CLVOL 47.29 m3 51.49 m3 Cost Estimates CONSTRUCTION COST 861,426.12 Dollars 962,038.28 Dollars OPERATIONAL COST 7,382.96 Dollars 8,245.27 Dollars Fuel Cost 6,771.74 Dollars 6,787.91 Dollars Cost Function CMF 86.15 ($/m3) 66.72 ($/m3) 41. The optimum vessel appears to be less than optimum when a l l the costs are considered, but the cost function is a measure of the ship's earning capability. It i s evident from Table IV, that by increasing the vessel design variables by less than 10 percent, and decreasing the speed by approximately 5 percent, the increase i n fuel cost is much less than the increase in fish hold capacity, hence the lower cost function value. This indicates that a speed reduction i s a very effective way to increase economic efficiency. b^fftoo ftr** / 7.3 Effect of Rising Fuel Costs on Optimum Design The price of fuel used in the optimization was 38.5 cents/ltr. which is the current price of fuel for fishing vessels. The program was run using fuel prices of 50 cents/ltr., 77 cents/ltr. (double the existing price) and $1.00 per l t r . to determine i f and how the vessel parameters would change with fuel price increases. When fuel price was set at 50 cents/ltr. the vessel dimensions did not change from the optimum at 38.5 cents/ltr. but the speed of the vessel was reduced. At 77 cents/ltr. the vessel dimensions changes, producing a finer ship, which is expected. When price was increased to $1.00 per l t r . the optimum vessel dimensions were the ame as the 77 cents/ltr. optimum but the vessel speed was reduced. Table V shows the vessel paramters for each price change. A possible explanation of the results shown in Table V may come from the definition of the optimum vessel in terms of the optimization c r i t e r i a . As stated previously the optimum set of design variables defining the ship would be those that resulted in the lowest fish hold capacity (FHC) to 42. Table V Fuel Price Increase Effects FUEL PRICE (cents/ltr.) 38.5 38.5 50.0 77.0 100.0 DESIGN VARIABLES EASTWARD HO OPTIMUM 1 OPTIMUM 2 OPTIMUM 3 OPTIMUM 4 XI (m) 29.26 30.395 30.395 31.247 31.247 X2 (m) 8.894 9.704 9.704 9.179 9.179 X3 (m) 2.926 2.948 2.948 2.532 2.532 X4 .8241 .874 .874 .8274 .8274 X5 .5251 .544 .544 .5135 .5135 X6 1.0206 .964 .932 .9188 .8915 dollar expended ratio. By using this c r i t e r i a the fuel price is taken as part of the total annual cost to operate the vessel, which may cause the optimization to be less sensitive to small price changes. Since the optimum vessel i s the ship with the lowest merit function, the effect of reducing speed for small changes in fuel price appers to be more cost effective than changing the entire vessel. For large changes i n fuel price (i.e. double) i t is apparent that a finer, shallower vessel is required to reduce resistance and increase fuel savings. Fishing vessels designers of the future (when fuel prices increase drastically) w i l l therefore have to be more concerned with the hydrodynamic efficiency of the hulls. 7.4 Percent Fish Hold Capacity Required to Break Even The optimum fishing vessel was derived independent of the type of f i s h caught. This section has therefore been included to determine what percentage of the fish holds must be f i l l e d in order for the owner to realize a profit. 43. This study has been centered around a trawler design which means the the only type of fish that i t is allowed to catch are bottom dwellers lik e cod, sole, and herring, etc. The season i s very short for herring, so the bulk of the trawler's catch w i l l most likely be bottom f i s h . The density of bottom f i s h i s approximately 7 9 0 kg/m3 while that of herring is approximately 9 3 0 kg/m3 [ 1 3 ] . Since the cost function is independent of the type of catch, i t i s easy to derive the price per pound of fish required when the holds are completely f u l l . The price per pound of f i s h using the optimized cost function i s : CMF Price Per Lb. = , ._. x . 4 5 3 7 $/lb. Density(Fish) Therefore the price per pound required for bottom fish with holds completely f u l l i s : Bottom Fish = 6 ^ ' Q 2 X . 4 5 3 7 = . 0 3 8 3 $/lb. For Herring: Herring = 6 ^ Q 2 X . 4 5 3 7 = . 0 3 2 6 $/lb. Bottom fish include rock code at 1 4 . 5 cents/lb, sole at 25 cents/lb and cod at 16 cents/lb [ 2 5 ] . Since trawling operations are indiscrimminant about the type of fish i t is assumed that the average price of the fish in the return to the fisherman, therefore the price per pound paid to the fisherman i s : 1 8 . 4 2 cents/lb for bottom fis h . Herring is a different case as the price dictated by the market i s very volatile, i n the 1 9 8 4 Herring Roe season the average price per lbs paid to the fisherman was 55 cents/lb [ 2 5 ] . 44. Table VI shows the price per pound of fis h required for various percentages of fish hold capacity for the optimum vessel. Table VI Fish Hold Capacity Price Require ($/lb) % F u l l Bottom Fish Herring 10 .3832 .3255 20 .1916 .1623 30 .1277 .1085 40 .0958 .0814 50 .0766 .0651 60 .0639 .0543 70 .0547 .0465 80 .0479 .0407 90 .0426 .0362 It is evident from Table VI that the owner of the Optimum vessel should f i s h herring a l l the time as with the holds only 10% f u l l he realized a profit approximately 22.5 cents per pound. Fishing for bottom fi s h the owner must have his holds at least 25 percent f u l l at a l l times to break even. As stated previously the herring season is very short therefore the owner would increase profits by fishing for the assumed 225 days at sea. 45. CHAPTER 8 DISCUSSION OF FUTURE TRENDS IN FISHING VESSEL DESIGN Ship designers [10] and ship builders [23] were canvassed to deter-mine their thoughts about future developments in fishing vessel design, in order to have a 'bench mark' by which to measure the program effectiveness. Appendix B i s an example of the questionnaire used to discuss fishing vessel trends. The trends predicted by this study follow similar trends to those predicted by the designers. This optimization indicates that as fuel price increases the vessel that Is produced has a finer hull form than that of an existing vessel, which is what was expressed by the local industry. Both the designers and builders feel that the existing Pacific Coast Fishing Fleet vessels are too big and that a well designed 55-60 f t . ship would be best for the industry. It was evident that more time would have to be spent developing fishing vessels with finer lines and good fish hold capacities. Future hulls would be displacement type, constructed of steel, which indicate that the assumption of steel construction used in this study w i l l be acceptable for vessels which are to replace the fleet. Future vessels w i l l be similar to existing vessels except that they w i l l be more hydrodynamically ef f i c i e n t , and the only dimension that would probably change would be the L/B ratio. Designers feel that diesel engines w i l l be the predominant propulsion prime mover, but feel that more time should be spent in allocating space so that separate auxiliary machinery can be used as i t i s more economical. The local industry Indicated that in future a fisherman's profit w i l l be determined by quality of fish rather than quantity therefore have predicted 46. that most vessels w i l l be f i t t i n g with refrigerated sea water systems for cooling the holds. It was also stated that the Pacific Coast Fishery should look at large mounted processing plants to reduce transit time for the fisherman, thereby reducing fuel costs. As evidenced by Table III this program can work for vessels in the 50-60 f t . range and therefore could be of use in future developments of the vessels required for the industry. The author feels that this optimization program is a f a i r l y accurate tool to be used at the preliminary design stage. Using this method the designer can quickly recommend the best dimensions for a vessel to conduct fishing operations. 8.1 Recommendations for Future Work It i s f e l t that to have a truly dependable preliminary design tool, the following areas should be developed. 1) The engine selection routine should have a l l makes and models of engines used in fishing vessels. 2) A study of the actual centers of gravity of components should be made, to accurately determine the st a b i l i t y of the vessel. 3) A comprehensive propeller design technique should be implemented. 4) The optimization program should be incorporated into a program like 'Spiral' to be able to calculate the hydrostatics, cross curves of st a b i l i t y , strength, and also l o f t the vessel. This would make the program a very effective preliminary design tool. 5) The net resistance should be developed and incorporated into the fishing scenario. 47. 6) The fishing simulation should be made more flexible, so the vessel can travel at various speeds for different periods of time. This means a variable scenario, as opposed to a fixed scenario. 7) A better or more accurate method to determine construction and operating costs should be looked into. Figure 4. Steel Weight Vs. Length Changes. 300 OH 1 1 r 1 1 1 15 20 25 30 35 40 LENGTH (m) Figure 5. West Coast Beam Vs. Depth. Legend A ACTUAL DATA x REGRESSION FIT DEPTH (m) Figure 6. West Coast Beam Vs. Length. Legend A ACTUAL DATA x REGRESSION FIT 10 20 â€” r -30 â€”Tâ€” 40 50 60 LENGTH (m) Figure 7. Cost Function Vs. Length. Figure 9. Cost Function Vs. Draft. U l Figure 12. Fuel Cost Vs. Length. 8000-1 6600 24 26 28~" 30 32 CHANGES IN LENGTH 34 Figure 13. Fuel Cost Vs. Beam. Figure 14. Fuel Cost Vs. Draft. 6600 n 6000 H 2.2 2.4 T 1 r 2.6 2.8 3 CHANGES IN DRAFT Figure 15. Fuel Cost Vs. Midship Coefficient. 6100 H 1 â€”â€”I 1 â€”i 1 1 0 70 0.76 0.80 0.86 0.90 0.96 1 MIDSHIP COEFFICIENT Figure 16. Fuel Cost Vs. Prismatic Coefficient. 8000-1 h-co O (J _ J U J 3 7600 7000 H 6600 6000 -r 0.46 0.60 0.66 0.80 PRISMATIC COEFFICIENT 0.66 61. REFERENCES 1. S.M. Calisal and 0. Fung; "Fuel Optimization Studies for Fishing Vessels". In print. 2. R.D. Murphy, D.J. Sabat and R.J. Taylor; "Least Cost Ship Characteristics by Computer Techniques". Marine Technology, Vol. 2, pp. 174-202, Society of Naval Architects and Marine Engineers, New York, N.Y. , 1965. 3. B. Paterson; "Least Cost Criterion Applied to Fishing Vessels". In print. 4. P. Mandel and R. Leopold; "Optimization Methods Applied to Ship Design". Transactions, Vol. 74, pp. 477-521, Society of Naval Architects and Marine Engineers, New York, N.Y., 1966. 5. L.K. Kupras; "Analysis of Main Dimensions, Displacement, Block Coefficient and Speed of a Stern Factory Trawler". FAO, Vol. 2, 1965. 6. C. Hamelin; "An Optimum Trawler for Groundfish: Design Study". U.S. Department of Commerce, National Marine Fisheries Service, Springfield, VA, 1971. 7. B. Mcllwaine; "Private Communications". Chief of Development, Department of Fisheries and Oceans, Vancouver, B.C., 1982-1984. 8. "Caterpillar Marine Engine Manuals". Steveston Marine, Richmond, B.C., 1983. 9. W.B. Wilson; "Fishing Vessel Design Curves". NOAA Data Buoy Center, NTSL, Mississippi, 1980. 10. R.F. Allan; "Private Communications". Robert Allan Naval Architects. 500 - 1380 Burrard St., Vancouver, B.C., 1984. 11. A. van Oortmerssen; "A Power Prediction Method and it s Application to Small Ships". Netherlands Ship Model Basin, Wagingen, The Netherlands, 1970. 12. S.M. Calisal and 0. Fung; "Resistance Comparison for Fishing Vessels". In print. 13. M.F.C. Santarelli; "Preliminary Determination of Main Characteristics of Fishing Vessels". Lecture Notes for Sixth Wegemt School, Fishing Vessel Technology, Madrid, Spain, 1982. 14. Comstock; "Principles of Naval Architecture". Society of Naval Architects and Marine Engineers, New York, N.Y., 1969. 15. Preliminary Steel Weight Data, Department of Fisheries and Oceans, Halifax, N.S., 1983. 62. 16. M.L. Frew; "Western Fisheries". 1132 Hamilton Street, Vancouver, B.C., 1970-1982. 17. R. Read and R. Latorre; "Fishing Vessel Intact Stability Criteria and Compliance Due to Variations in Vessel Dimensions". International Conference on Design, Construction and Operation of Commercial Fishing Vessels, Melbourne, Florida, 1984. 18. International Maritime Consultive Organization; "Safety of Fishing Vessels". Torremolos Conference, 1977. 19. J.R. Amy, R.E. Johnson and E.R. Miller; "Development of Intact Stability Criteria for Towing and Fishing Vessels". Transactions, Vol. 84, pp. 75-114, Society of Naval Architects and Marine Engineers, New York, N.Y., 1976. 20. Canadian Imperial Bank of Commerce, Industrial Development Bank; "Private Communications". Vancouver, B.C., 1984. 21. B.D. Smith; "Private Communications". Marine Surveyor, Vancouver, B.C., 1984. 22. Openshaw Insurance & Lloyds; "Private Communications". Insurance Brokers, Vancouver, B.C., 1984. 23. MacLaren; "Private Communications". President Allied Shipbuilding Co. Ltd., North Vancouver, B.C., 1984. 24. B. Henry; "Private Communications". Wilkinson Steel of Canada, Vancouver, B.C., 1984. 25. J. Kemp; "Private Communications". B.C. Packers, Steveston, B.C., 1984. 63. APPENDIX A COMPARISON OF STEEL WEIGHT ESTIMATES TO WEST COAST INVOIDED STEEL WEIGHTS The decision to use the average of the steel weight estimate formula-tions was based on the percentage error between known invoiced steel weights (for West Coast vessels) and the estimates. The following tables show the actual steel weight, for 3 classes of vessel constructed at Al l i e d Shipbuilding in North Vancouver, as well as the estimates and percent error. Class I Vessels Dimensions: L = 16.69 m B = 5.03 m D = 2.44 m Formulation Steel Weight % Error Actual 27.28 tonne -East Coast 41.66 tonne 52.71 Santarelli 1 Santarelli 2 Santarelli 3 25.68 tonne 40.86 tonne 57.73 tonne 5.87 49.78 111.62 Wilson 24.05 tonne 11.84 Superstructure =0 Superstructure = 50 Superstructure = 100 Since there are no vessels constructed without superstructures the f i r s t Santarelli formulation is neglected. The third Santarelli formula-tion i s neglected because of too high of error. Therefore: Avg. Steel Weight Estimate = 35.52 tonne Avg. Percent Error = 30.22 percent 64. Class II Vessels Dimensions: L = 23.17 m B = 7.01 m D = 3.20 m Formulation Steel Weight % Error Actual 77.85 tonne -East Coast 79.00 tonne 1.48 Santarelli 2 80.05 tonne 3.40 Wilson 61.50 tonne 21.00 Avg. Steel Weight Estimate = 73.52 tonne Avg. Percent Error = 5.57 percent Class III Vessels Dimensions: L = 35.36 m B = 9.75 m D = 4.75 m Formulation Steel Weight % Error Actual 216.35 tonne -East Coast 328.67 tonne 51.92 Santarelli 2 200.2 tonne 7.47 Santarelli 3 225.08 tonne 4.04 Wilson 198.23 tonne 8.38 The East Coast formulation i s neglected because of the high error. Therefore: Avg. Steel Weight Estimate = 207.84 tonne Avg. Percent Error = 3.93 percent 65. Proper estimation of steel weight i s very important as the sum of a l l the weight groups are subtracted from the ship's calculated displacement, to check that there i s positive buoyancy, I.e. so the ship w i l l float. It is also used to determine the construction and operating costs of the vessel. 66. APPENDIX B QUESTIONNAIRE USED IN DISCUSSION WITH LOCAL INDUSTRY WITH REGARD TO FUTURE TRENDS IM FISHING VESSEL DESIGN Date: Name of Designer/Builder: Most Recent Fishing Vessel Design: Section I Trends in Hull Forms 1. What type of hull w i l l be used in fishing vessels of the future? Displacement, semi-planing, planing. 2. Will hulls of the future have f u l l middle bodies? 3. What type of stern w i l l be used? 4. Will there be changes in the ranges of coefficients used for vessels of the future? 5. What materials w i l l be used in future vessel construction? Section II Power Requirements 1. What type of powering w i l l be used in the vessels? (Diesel, gasoline, outboard-inboard.) 2. Will separate auxiliaries be used? 3. Will the ships have refrigerated sea water systems? Section III General 1. What are your impressions about future vessel requirements? 67. APPENDIX C PROBLEMS EXPERIENCED WITH THE DESIGN AND OPTIMIZATION The following problems have been experienced when using the programs. 1. In using only Caterpiller engines in the engine selection, many vessels are rejected because no suitable engine can be found. 2. The resistance subroutine i s very dependent on prismatic coefficient, which when varied too much causes the resistance result to be erroneous. 3. The fishing simulation causes problems, when the engine selected i s at either end of the scale, by not being able to match engine RPM to the required horsepower when the displacement of the vessel i s increased.
- Library Home /
- Search Collections /
- Open Collections /
- Browse Collections /
- UBC Theses and Dissertations /
- Fishing vessel optimization : a design tool
Open Collections
UBC Theses and Dissertations
Featured Collection
UBC Theses and Dissertations
Fishing vessel optimization : a design tool Bower, Thomas Charles 1985
pdf
Page Metadata
Item Metadata
Title | Fishing vessel optimization : a design tool |
Creator |
Bower, Thomas Charles |
Publisher | University of British Columbia |
Date Issued | 1985 |
Description | Rising fuel costs and decreased catch sizes have reduced the fishing vessel owners profit margin. This has caused the owners to try to find methods that reduce the costs of their operations. In this thesis a tool which can be used by fishing vessel designers, and operators, is developed for use at the preliminary ship design stage. It is used to determine the best fishing vessel parameters for a given operational scenario found on the West Coast of Canada. |
Subject |
Fishing boats |
Genre |
Thesis/Dissertation |
Type |
Text |
Language | eng |
Date Available | 2010-05-27 |
Provider | Vancouver : University of British Columbia Library |
Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
DOI | 10.14288/1.0096296 |
URI | http://hdl.handle.net/2429/25082 |
Degree |
Master of Applied Science - MASc |
Program |
Mechanical Engineering |
Affiliation |
Applied Science, Faculty of Mechanical Engineering, Department of |
Degree Grantor | University of British Columbia |
Campus |
UBCV |
Scholarly Level | Graduate |
AggregatedSourceRepository | DSpace |
Download
- Media
- 831-UBC_1985_A7 B68.pdf [ 2.67MB ]
- Metadata
- JSON: 831-1.0096296.json
- JSON-LD: 831-1.0096296-ld.json
- RDF/XML (Pretty): 831-1.0096296-rdf.xml
- RDF/JSON: 831-1.0096296-rdf.json
- Turtle: 831-1.0096296-turtle.txt
- N-Triples: 831-1.0096296-rdf-ntriples.txt
- Original Record: 831-1.0096296-source.json
- Full Text
- 831-1.0096296-fulltext.txt
- Citation
- 831-1.0096296.ris
Full Text
Cite
Citation Scheme:
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>
Our image viewer uses the IIIF 2.0 standard.
To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0096296/manifest