optimizing deployment of shiploaders at bulk export terminal

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

OPTIMIZING DEPLOYMENT OF SHIPLOADERS AT BULK

EXPORT TERMINAL

By Lal C. Wadhwa1

ABSTRACT: Bulk commodity export terminals are equipped with very expensive infrastructure. The massivecapital outlay also is accompanied by large operating costs associated with major equipment such as shiploaders,reclaimers, and unloading stations. The port management is always interested in optimizing the deployment ofits infrastructure with a view to minimizing the terminal’s operating costs. This study deals with finding anoptimal solution to an interesting situation where using one shiploader results in unacceptable ship waiting timesand a high level of demurrage, whereas continuous deployment of two shiploaders results in inefficiency andhigh operating costs. This paper describes an approach for developing a strategy that considers a trade-offbetween the ship waiting cost and the cost of deploying the additional shiploader and results in the optimaldeployment of resources. The approach is an integration of simulation, scenario building, and economic fun-damentals. An actual bulk export terminal in Australia is used to demonstrate the applicability of this approachin assisting the port management in rational decision making.

INTRODUCTION

Outloading Infrastructure

The outloading infrastructure of a bulk export terminal maycomprise stockpiles or storage sheds, reclaimers, surge bins,outloading strings, and shiploaders. It may receive its com-modities from several sources and store them in stockpiles orsheds at the terminal.

The terminal in question has four rows of stockpiles with atotal capacity of >2,000,000 t. Stockpile configurations can bealtered, but currently there are 33 individual stockpiles withcapacities ranging from 13,300 to 147,000 t. Stockpiles con-tain coal from different mines and may have different prod-ucts. The terminal has six reclaimers and two outloadingstrings, each with its own surge bins. There are two shipload-ers. In fact, there are two parallel outloading systems, eachwith a rated capacity of 7,200 t/h. The terminal services shipsrange from handy size of 25,000 dead weight tonnage (DWT)to very large carriers of 200,000 DWT, with an average of87,100 DWT during 1998–1999. The outloading system at theterminal is represented schematically in Fig. 1.

Operational Procedure

When a ship is to be loaded, two reclaimers, if available,reclaim the commodity from designated stockpiles containingthe requested product. It is loaded by a shiploader through anoutloading system consisting of a surge bin and several kilo-meters of belt conveyor. A ship is loaded with one shiploaderonly, although each shiploader can work on either of the twoberths. Normally the newer shiploader (SL2) is deployed forloading ships irrespective of the berth at which the ship is tobe loaded. The second shiploader (SL1) is deployed only whenSL2 is under maintenance or two ships have to be serviced atthe same time. The shiploaders are not dedicated to a berthand can travel to either berth. However, shiploaders cannotcross and SL2 will always remain to the left of the older ship-loader (SL1). It is also the policy of the management of theterminal not to use both shiploaders for loading a single ship.Only very few ships (cape-size carriers loading a single prod-

1Head, Civ. and Envir. Engrg., James Cook Univ., Queensland 4811,Australia. E-mail: [email protected]

Note. Discussion open until May 1, 2001. To extend the closing dateone month, a written request must be filed with the ASCE Manager ofJournals. The manuscript for this paper was submitted for review andpossible publication on January 22, 1999. This paper is part of the Jour-nal of Waterway, Port, Coastal, and Ocean Engineering, Vol. 126, No.6, November/December, 2000. qASCE, ISSN 0733-950X/00/0006-0297–0304/$8.00 1 $.50 per page. Paper No. 20110.

JOURNAL OF WATERWAY, PORT, COASTA

J. Waterway, Port, Coastal, Oce

FIG. 1. Outloading System

uct) may be able to accept two loaders simultaneously. Thisoption is not considered in this study.

Problem Definition

At the rated capacity of 7,200 t/h, one shiploader canpractically handle a throughput of >30 million t per annum(MTPA) under acceptable performance levels. Two shiploaderswill be gainfully deployed when the throughput is in excessof 42 MTPA. However, at the present and in the near future,the throughput is expected to range around 30–35 MTPA. Thismeans that the continuous deployment of two shiploaders maynot be fully justified due to lower utilization rates and, con-sequently, idle crews.

Shiploader SL2 is deployed for the loading of all ships, andSL1 is assigned only while SL2 is undergoing maintenance.This means that one shiploader is always available with crewassigned. This policy is based on the conviction that the costof continually manning two shiploaders could be prohibitive.This may not, however, be an optimal practice as queues mayform, waiting time and turnaround time of ships may increase,and demurrage may be too high. To relieve congested situa-tions, SL1 may have to be deployed selectively.

Objectives and Approach

The approach to the development of an optimal operationalregime is to simulate alternative shiploader deployment strat-egies. These strategies may be represented in the form of rulesunder which the second shiploader may be brought into use.The objective is to recommend a desirable strategy that willoffset the additional costs of operating the second shiploader(which includes crew and operating costs) by reducing de-murrage. This is achieved by

• Simulating the performance of the outloading operationsby using the port simulation model developed at JamesCook University, Queensland, Australia, for the exportterminal

L, AND OCEAN ENGINEERING / NOVEMBER/DECEMBER 2000 / 297

an Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

TABLE 1. Anchorage Queue for Deployment of Second Ship-loader (SL1)

Strategy(1)

Anchoragequeue

(2)Remarks

(3)

A 0 As soon as second ship berthsB 1 Two ships berthed, one in queueC 2 Two ships berthed, two in queueD 3 Two ships berthed, three in queueE Infinite No deployment of second shiploader (except

when SL2 is undergoing maintenance)

• Analyzing the relationship between the frequency and du-ration of deployment of the second shiploader, demurrage,and throughput for each deployment strategy

• Recommending the optimal strategy from among thoseconsidered based on economic and operational efficiencyconsiderations.

SIMULATION ENVIRONMENT

Simulation Model

The integrated front- and back-of-port model developed forthe export terminal by the School of Engineering at JamesCook University has been deployed in this study. The modelis programmed in a special-purpose language, ARENA. Anearlier version of this model was used to evaluate alternativestrategies for planning the operations of this terminal (Wadhwa1992).

Data and information relating to the proportion of entitle-ment by various users of the terminal (tonnage per annum),the ship generation stream, the durations and frequency of portclosures, maintenance regimes for the outloading strings andberths, etc., are provided by the management of the terminaland used as model inputs. The proportion of ships carryingblended cargo was ascertained for each ship category from theavailable data.

Formulating Strategies

It is hypothesized that the deployment of the second ship-loader should be triggered by the state of the system. In anysystems study, an action is initiated when the system statereaches a certain threshold or critical level. In this study, thestate that should logically trigger the deployment of the secondshiploader is taken to be the anchorage queue length.

Although this approach to formulating alternative strategiescan lead to a continuum of possible deployment strategies, thefollowing five strategies are selected for simulation and anal-ysis (Table 1). The deployment of the second shiploader ceasesas soon as the condition that resulted in its deployment nolonger exists.

Shiploader Maintenance Schedule

The maintenance schedule for SL2 is as follows:

• 12 h every 1,000,000 t• 36 h every 4,000,000 t• 132 h every 8,000,000 t

Table 2 shows a sample of the above-mentioned policy. Theschedule has an 8,000,000-t cycle, and two cycles are dis-played in Table 2. The loader is allowed to complete loadingthe ship before undergoing maintenance. Based on the actualthroughput, the maintenance cycles will continue as required.Because SL1 is not deployed continuously, its maintenance isnot modeled. It is presumed that maintenance on SL1 will be

298 / JOURNAL OF WATERWAY, PORT, COASTAL, AND OCEAN ENGI

J. Waterway, Port, Coastal,

TABLE 3. Allowed Laytime

DWT(t)(1)

Rate(t/day)

(2)

<45,000 15,00045,000 < DWT < 60,000 20,00060,000 < DWT < 75,000 25,00075,000 < DWT < 100,000 35,000

100,000 < DWT < 125,000 40,000>125,000 45,000

TABLE 2. Maintenance Schedule for Shiploader SL2

Load(t)(1)

Maintenance(h)(2)

1,000,000 122,000,000 123,000,000 124,000,000 365,000,000 126,000,000 127,000,000 128,000,000 1329,000,000 12

10,000,000 1211,000,000 1212,000,000 3613,000,000 1214,000,000 1215,000,000 1216,000,000 132

carried out during its nonallocated period and therefore willalways be available for deployment when required.

Range of Throughput Simulated

Throughputs ranging from 28 to 33 MTPA, have been sim-ulated for each strategy.

Despatch/Demurrage

One of the most relevant measures of a terminal’s opera-tional performance is despatch/demurrage. This economic in-dicator is a function of the time spent by the ship in the port.If a ship’s actual laytime is less than the allowed laytime de-rived rationally by an established formula, the terminal is cred-ited with ‘‘despatch’’ (extra income); however, if the ship’stime in port exceeds the allowed laytime, the terminal incursdemurrage. The rate of demurrage is twice the rate of despatch,and both are charged on an hourly basis. Despatch results inextra income for the port, whereas demurrage costs the portin terms of reduced payment.

The allowed laytime is calculated by dividing the quantityof coal loaded by a rate that is a function of the ship’s DWT.This is shown in Table 3. The actual laytime is given by

(queuing time 2 12 h) 1 (completion of loading 2 start of loading)

SIMULATIONS OF DEPLOYMENT STRATEGIES

This study required the simulation of five strategies each forsix levels of throughput. Therefore 30 full-scale simulationshave been carried out for a period of 10 years and the averageannual statistics have been derived by ignoring the outputs forthe first 2 years to overcome the effect of transient fluctuations.

The key statistics that have been considered in evaluatingalternative strategies include

• Demurrage• Frequency of deployment of second shiploader

NEERING / NOVEMBER/DECEMBER 2000

Ocean Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

TABLE 5. Summary of Results for Strategy B

Throughput(MTPA)

(1)

Frequency ofdeployment

(number/year)(2)

Hoursdeployed(h/year)

(3)

Despatcha

(dollars/t)(4)

28.21 55 887 0.068329.27 68 998 0.058130.52 79 1,170 0.051531.50 83 1,228 0.040532.82 97 1,450 0.028633.53 105 1,582 0.0178

aNo demurrage is represented here.

TABLE 4. Summary of Results for Strategy A

Throughput(MTPA)

(1)


(number/year)(2)


(3)

Despatcha

(dollars/t)(4)

28.28 110 1,648 0.066829.23 123 1,828 0.062730.60 135 2,038 0.05331.98 144 2,155 0.040733.22 161 2,442 0.0134

aNo demurrage as represented here.

• Total number of hours for which second shiploader hasto be deployed in a year

Strategy A: Deploy Second Shiploader As Soon AsSecond Ship Is Berthed

This strategy requires frequent deployment of the secondshiploader. The results of simulating this strategy are sum-marized in Table 4. The number of times and the hours forwhich the second shiploader is used are the maximum amongall strategies considered. However, as expected, the despatchis maximum for this strategy. The relatively high despatch isachieved at a significant cost in deploying the second ship-loader. Basically this relates to crew costs for the entire out-loading string, including the shiploader.

In this strategy, the despatch decreases from a maximum of$0.0668 at 28 MTPA to just $0.0134 at 33 MTPA. The numberof times for which the second shiploader had to be deployedis shown to vary from 110 to 161. The second shiploader isused for 1,648 h in the case of 28 MTPA and increases to2,442 h for 33 MTPA.

Strategy B: Deploy Second Shiploader When TwoShips Are Berthed and Another Ship Enters Queue

This strategy also requires frequent deployment of the sec-ond shiploader. The results of simulating this strategy are sum-marized in Table 5. The number of times and the hours forwhich the second shiploader is used are quite substantial. Thedespatch is still high but generally lower than for Strategy A.

In this strategy, the despatch decreases from a maximum of$0.0683 at 28 MTPA to a minimum of $0.0178/t at 33 MTPA.The number of times for which the second shiploader has tobe deployed is substantially lower than Strategy A and isshown to vary from 55 to 105. The second shiploader is usedfor 887 h for 28 MTPA and increases to 1,582 h for a through-put of 33 MTPA.

Strategy C: Deploy Second Shiploader When TwoShips Are Berthed and Second Ship JoinsAnchorage Queue

The results of simulating this strategy are summarized inTable 6. The despatch decreases from $0.057/t at 28 MTPA to

JOURNAL OF WATERWAY, PORT, COAST

J. Waterway, Port, Coastal, Oc

TABLE 8. Summary of Results for Strategy E

Throughput(MTPA)

(1)

Despatch/demurragea

(dollars/t)(2)

28.03 0.010629.13 20.012830.06 20.054531.00 20.10932.01 20.195432.90 20.351133.37 20.5172

aDemurrage is represented with minus sign.

TABLE 7. Summary of Results for Strategy D

Throughput(MTPA)

(1)


(number/year)(2)


(3)

Despatch/demurragea

(dollars/t)(4)

28.03 13 163 0.036229.09 16 256 0.036530.09 27 364 0.014631.05 28 476 0.001331.99 34 398 20.006233.09 48 739 20.0351


TABLE 6. Summary of Results for Strategy C

Throughput(MTPA)

(1)


(number/year)(2)


(3)

Despatch/demurragea

(dollars/t)(4)

28.21 32 468 0.05729.68 42 671 0.04730.60 44 657 0.048631.77 44 766 0.045532.59 51 737 0.026533.20 59 914 20.0012


a slight demurrage of $0.002 at 33 MTPA. The number oftimes for which the second shiploader has to be deployed isshown to vary from 32 to 59. The second shiploader is usedfor 468 h for 28 MTPA and increases to 914 h for a throughputof 33 MTPA.

Strategy D: Deploy Second Shiploader When TwoShips Are Berthed and Anchorage Queue Increasesto Three

The results of simulating this strategy are summarized inTable 7. The despatch decreases from a maximum of $0.0362/tat 28 MTPA to a demurrage of $0.0351/t at 33 MTPA. Thenumber of times for which the second shiploader has to bedeployed is shown to vary from 13 to 48. The second ship-loader is used for 163 h for 28 MTPA and increases to 739 hfor a throughput of 33 MTPA.

Strategy E: Do Not Deploy Second Shiploader ExceptWhen SL2 Is Undergoing Maintenance

This strategy requires minimal deployment of the secondshiploader. The results of simulating this strategy are sum-marized in Table 8. The number of times and the hours forwhich the second shiploader is used are the minimum of allstrategies considered. However, the demurrage is maximumunder this strategy. The high demurrage is a trade-off due tolimited deployment of the second shiploader. A despatch of

AL, AND OCEAN ENGINEERING / NOVEMBER/DECEMBER 2000 / 299

ean Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

$0.01/t at 28 MTPA rapidly drops to a demurrage of about$0.52/t at 33.4 MTPA. The frequency and duration of deploy-ment of SL1 under this strategy is only dependent on the main-tenance schedule of SL2. SL1 is deployed every time SL2undergoes maintenance. It is, therefore, not a function of thethroughput handled or of a specific strategy of deployment;hence, the frequency of deployment and the hours for whichthe second shiploader is deployed are not examined for thisstrategy.

INTERPRETATION OF SIMULATION RESULTS

Some obvious results from the simulation of alternativestrategies are that

• In moving from Strategy A to E, the deployment of thesecond shiploader reduces in frequency and total durationof deployment.

300 / JOURNAL OF WATERWAY, PORT, COASTAL, AND OCEAN ENGINEE

J. Waterway, Port, Coastal, Oce

• As the deployment of the second shiploader decreases, thedespatch decreases and demurrage levels increase.

• For the same strategy, demurrage increases as throughputincreases.

However, an in-depth analysis of simulation results (fivestrategies and throughput levels ranging from 28 to 33 MTPA)is required to fully appreciate the trade-offs involved to makea rational and logical recommendation.

Despatch/Demurrage

The effect of adopting alternative strategies on despatch/demurrage has been examined for various levels of throughput.The effect of increasing throughput on despatch/demurrage isshown for each strategy in Fig. 2, and the effect of each strat-egy is shown in Fig. 3 for various throughput levels. Fig. 2

FIG. 2. Effect of Throughput on Despatch/Demurrage for Various Strategies (1 cent = $0.01)

FIG. 3. Effect of Strategy on Despatch/Demurrage for Various Throughputs (1 cent = $0.01)

RING / NOVEMBER/DECEMBER 2000

an Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

clearly shows the merits of Strategies A–C at all throughputlevels, and Fig. 3 shows the inappropriateness of Strategy Eas throughput increases. It is obvious that as throughput in-creases, the strategy of deploying the second shiploader witha queue of one or two ships becomes more attractive. Thestrategy of deploying SL1 only in case of the maintenance ofSL2 (Strategy E) results in significant demurrage at higherthroughput.

Frequency of Deployment of Second Shiploader

The frequency of deployment of the second shiploader as afunction of throughput for each strategy is shown in Figs. 4and 5. Fig. 4 shows the rapid fall in the frequency of deploy-

JOURNAL OF WATERWAY, PORT, COASTAL

J. Waterway, Port, Coastal, Ocea

ment in going from Strategy A to B, but the rate of declinereduces when shifting from Strategy B to C and further on toD. Fig. 5 shows that there is a moderate increase in the fre-quency of deployment with an increase in throughput for anystrategy. It is obvious that the frequency of deployment forany strategy increases with an increase in throughput. As thestrategy shifts toward deploying the second shiploader withlonger queues, the frequency of deployment decreases.

Proportion of Ships Serviced by Second Shiploader

As throughput increases, the number of ships loaded alsoincreases. It is instructive to examine if the proportion of shipsthat are serviced by the second shiploader for each strategy

FIG. 5. Effect of Strategy on Frequency of Deployment for Various Throughputs

FIG. 4. Effect of Throughput on Frequency of Deployment for Various Strategies

, AND OCEAN ENGINEERING / NOVEMBER/DECEMBER 2000 / 301

n Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

TABLE 9. Proportion of Ships Serviced by Second Shiploader(%)

Throughput(MTPA)

(1)Strategy A

(2)Strategy B

(3)Strategy C

(4)Strategy D

(5)

28 30.1 15.0 8.8 3.629 32.4 17.9 11.1 4.230.5 33.8 19.8 11.0 6.832 34.7 19.5 11.3 8.233 37.4 23.5 13.7 11.2

also increases with the level of throughput. This proportionhas been determined for all throughput levels and strategies(Table 9) and follows a similar pattern to the frequency ofdeployment.



Total Duration of Deployment of Second Shiploader

The duration of deployment is expressed in terms of thenumber of hours for which the second shiploader is expectedto be deployed under any strategy. Figs. 6 and 7 show theduration for which the second shiploader is expected to bedeployed under each strategy for various throughput levels.These figures reinforce the pattern of variations in the durationof deployment with increasing throughput and for variousstrategies similar to the effect on the frequency of deployment.

Average Deployment Duration

The average number of hours for which the second ship-loader is deployed is remarkably similar for all throughputsand strategies. This figure has been found to be around 15 62 h for each deployment. There is no pattern to suggest a

FIG. 7. Effect of Strategy on Hours of Deployment of Second Shiploader for Various Throughputs

FIG. 6. Effect of Throughput on Hours of Deployment of Second Shiploader for Various Strategies


n Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

TABLE 12. Effectiveness of Deployment of Second Ship-loader

Throughput(MTPA)

(1)

Increase indeployment

(h)(2)

Increase indespatch(dollars/t)

(3)Elasticity

(4)

28 1,485 0.0306 0.2129 1,572 0.0265 0.1730.5 1,618 0.048 0.3032 1,575 0.0472 0.3033 1,703 0.0485 0.28

TABLE 11. Effect of Strategy on Deployment and Despatch/Demurrage

Throughput(MTPA)

(1)

Increase inDeployment (h)

FromStrategy D

(2)

ToStrategy A

(3)

Increase in Despatch(dollars/t)

FromStrategy D

(4)

ToStrategy A

(5)

28 163 1,648 0.0362 0.066829 256 1,828 0.0365 0.06330.5 420 2,038 0.005 0.05332 580a 2,155 20.0062 0.04133 739 2,442 20.0351 0.0134

aInterpolated.

TABLE 10. Effect of Increase in Throughput on Deploymentand Despatch/Demurrage

Strategy(1)

Increase inDeployment (h)

From(2)

To(3)

Decrease in Despatch(dollars/t)

From(4)

To(5)

A 1,648 2,442 0.0668 0.0134B 887 1,500 0.0683 0.025C 468 914 0.057 20.0012D 163 739 0.0362 20.0351

systematic effect due either to the throughput or the strategyon the average duration of deployment of the second ship-loader.

ELASTICITY OF DESPATCH/DEMURRAGE WITHRESPECT TO SHIPLOADER DEPLOYMENT

When the deployment of the second shiploader is reduced,the despatch decreases and demurrage increases. An in-depthanalysis of the relative effect of changes in the number ofhours of deployment of the second shiploader for each strategywas undertaken.

Effect of Increase in Throughput

As throughput increases from 28 to 33 MTPA, the hours ofdeployment increase and despatch decreases (or demurrage in-creases). This is shown in Table 10. The increase in the num-ber of hours deployed is not able to hold the despatch at thevalue achieved at the throughput level of 28 MTPA.

Effect of Strategy on Effectiveness of Deployment

In moving from Strategy D toward Strategy A, the durationof deployment of the second shiploader increases from 163 to1,648 h/annum for a throughput of 28 MTPA. This is accom-panied by an increase in despatch from $0.0362/t to $0.0668/t. This is shown in Table 11 along with similar data for otherthroughput levels.

The effectiveness of deployment of the second shiploader

JOURNAL OF WATERWAY, PORT, COASTAL


at each throughput level may be inferred from the elasticityof the strategy, which is calculated as the increase in despatch(in dollars per ton) per 100 h of increase in the deploymentof the second shiploader. Table 12 shows that increasing thehours of deployment by switching strategies (e.g., from D to-ward A) brings much higher returns at throughputs of 30MTPA or higher. This is a significant result from the detailedanalysis of simulation results.

DISCUSSION AND CONCLUSIONS

Realizing that the current throughput level at the terminaldoes not warrant the continuous deployment of two shipload-ers, it has been premised that

• The new shiploader (SL2) be used as the primary loadingfacility

• The previous shiploader (SL1) be deployed whenever SL2is undergoing maintenance

• Based on an appropriate strategy triggered by the systemstate, both SL1 and SL2 be deployed to load two shipssimultaneously

Four alternative deployment strategies have been formulatedin addition to the strategy of not deploying the second ship-loader except during maintenance of SL2 (Strategy E). Theseare

• Deploy SL1 as soon as the second ship is berthed (Strat-egy A).

• Deploy SL1 as soon as the third ship arrives in the system(i.e., two ships are berthed and the third has to join thequeue—only one ship in the queue) (Strategy B).

• Deploy SL1 as soon as the fourth ship arrives in the sys-tem (i.e., two ships are berthed, there is one ship alreadyin the queue, and the new ship joins the queue whoselength becomes two ships) (Strategy C).

• Deploy SL1 as soon as the fifth ship arrives in the system(i.e., two ships are berthed, there are two ships already inthe queue, and the new ship joins the queue whose lengthbecomes three ships) (Strategy D).

These strategies have been simulated using throughput levelsbetween 28 and 33 MTPA. Some significant results of thisstudy include the following:

• The rate of reduction in the frequency of deployment ofthe second shiploader decreases as one moves toward thestrategies of deployment with longer queue length.

• The proportion of ships serviced by the second shiploaderincreases with an increase in throughput and decreaseswith the move from Strategy A to Strategy D. This patternis similar to the frequency of deployment variations.

• The average number of hours for which the second ship-loader is deployed for each deployment is remarkablysimilar across all throughput levels and strategies. Theaverage value is 15 6 2 h for each deployment. Eachdeployment will require an average of two shifts of crew.

• As the throughput increases from 28 to 33 MTPA, thenumber of hours of deployment of the second shiploaderincreases by 500–800 h/annum while the despatch re-duces by about $0.05–$0.07/t for the range of strategiesconsidered.

• Increasing the hours of deployment by following alter-native strategies brings much higher returns ($0.003/t in-crease in despatch for every 100 additional hours of de-ployment per year) at throughput levels of 30 MTPA orhigher compared to a benefit of <$0.002/t at throughputs<30 MTPA. (The effectiveness of deployment is measured

, AND OCEAN ENGINEERING / NOVEMBER/DECEMBER 2000 / 303

n Eng. 2000.126:297-304.

Dow

nloa

ded

from

asc

elib

rary

.org

by

Uni

vers

ity o

f L

eeds

on

09/0

5/13

. Cop

yrig

ht A

SCE

. For

per

sona

l use

onl

y; a

ll ri

ghts

res

erve

d.

by the elasticity of deployment, which has been definedas the increase in despatch per 100 h of additional de-ployment of the second shiploader.)

It is recommended that the second shiploader need not be de-ployed except during maintenance of SL2 (Strategy E) at athroughput of 28 MTPA. This policy is, however, uneco-nomical for throughputs >30 MTPA. The benefits of deployingthe second shiploader are greater at higher throughputs, and astrategy of deploying the second shiploader when the secondship joins the anchorage queue while both berths are occupied(Strategy C) is quite appropriate.



ACKNOWLEDGMENTS

The continuing help of Manfred Schneider in the development of thisstudy, discussion of the methodology, and critical assessment of prelim-inary results is highly appreciated. Dominique Morel carried out the sim-ulations while working in the United Kingdom.

APPENDIX. REFERENCES

Wadhwa, L. C. (1992). ‘‘Planning operations of bulk loading terminalsby simulation.’’ J. Wtrwy., Port, Coast., and Oc. Engrg., ASCE, 118(3),300–315.


n Eng. 2000.126:297-304.

optimizing deployment of shiploaders at bulk export terminal

Documents