《系统工程》课程教学资源（英文文献）A System Dynamics Model for Port Operation System Based on Time, Quality and Profit

A System Dynamics Modelfor Port OperationSystem Based on Time, Quality and ProfitGuide Words: Port Operation System, System Dynamic Model, Time, Quality and Profit.Abstract: A novel and generic system-dynamic-model is proposed, which analyzes port operationsystem from perspectives of time, quality and profit. First, by studying the internal structure andoperation mechanism of port operation system, a sophisticated port operation system is defined anddivided into three subsystems: time, quality and profit. Later a system dynamic-model including threesub models is built via system dynamical methodology.Finally, an enterprise in Lianyungang Port isselectedasacasestudytoseekeffectivemethodstosolveissuesofportoperationsystemthroughthesimulation. The purpose of this study is to simulate the operational process at the port in order toprovide a useful tool to achieve three management goals: guarantying service time, improving qualityand reducing cost of port service. This research indicates that this model can be applicable to any portto analyze port operation system and provide support for the decision-making of port strategy and portpolicies.I.INTRODUCTIONIn the middle 1990s, China initiated the policy of “separate government functions from enterprisemanagement" in its port system. By 2001, China had implemented this policy in a full scale, with portenterprises becoming independent legal entities who assume sole responsibility for their own profitsand losses. The key to the development of port industry lies in better utilization of limited resourcesmore satisfying service level and quality, as well as increased profits. If port enterprises do not runwell, then ports will lose their original competition leverage such as customers, market, and profits,which will create a vicious cycle in port development. Therefore, port enterprises must be veryfamiliar with their operation systems and locate the crux of any problem in time and make adjustmentsaccordinglyinordertoeffectivelycontroland improvetheiroperationsystemsandenhancetheportcompetition.Currently, there are some papers studying the area of the port operation system (POS) or therelated fields. In [1], two different operational systems, referring to current and proposed, wereinvestigated and compared via a simulation model. Based on the comparison, issues of theperformance criteria were addressed and the way was found to reduce port terminal congestion andincrease thecapacityof terminal.Bottleneckpointswere created by analyzingloading/unloadingvehicles after complex port structures are investigated through simulation techniques in [2]. In [3]

A System Dynamics Model for Port Operation System Based on Time, Quality and Profit Guide Words：Port Operation System, System Dynamic Model, Time, Quality and Profit. Abstract：A novel and generic system-dynamic-model is proposed, which analyzes port operation system from perspectives of time, quality and profit. First, by studying the internal structure and operation mechanism of port operation system, a sophisticated port operation system is defined and divided into three subsystems: time, quality and profit. Later a system dynamic-model including three sub models is built via system dynamical methodology. Finally, an enterprise in Lianyungang Port is selected as a case study to seek effective methods to solve issues of port operation system through the simulation. The purpose of this study is to simulate the operational process at the port in order to provide a useful tool to achieve three management goals: guarantying service time, improving quality and reducing cost of port service. This research indicates that this model can be applicable to any port to analyze port operation system and provide support for the decision-making of port strategy and port policies. I. INTRODUCTION In the middle 1990s, China initiated the policy of “separate government functions from enterprise management” in its port system. By 2001, China had implemented this policy in a full scale, with port enterprises becoming independent legal entities who assume sole responsibility for their own profits and losses. The key to the development of port industry lies in better utilization of limited resources, more satisfying service level and quality, as well as increased profits. If port enterprises do not run well, then ports will lose their original competition leverage such as customers, market, and profits, which will create a vicious cycle in port development. Therefore, port enterprises must be very familiar with their operation systems and locate the crux of any problem in time and make adjustments accordingly in order to effectively control and improve their operation systems and enhance the port competition. Currently, there are some papers studying the area of the port operation system (POS) or the related fields. In [1], two different operational systems, referring to current and proposed, were investigated and compared via a simulation model. Based on the comparison, issues of the performance criteria were addressed and the way was found to reduce port terminal congestion and increase the capacity of terminal. Bottleneck points were created by analyzing loading/unloading vehicles after complex port structures are investigated through simulation techniques in [2]. In [3]

study on Port of Singapore indicates a combination of resources can help create a sustainableadvantageforthePort.Resourcesincluded supportivegovernmentpolicies,ample investment,andwell thought out operations and information technology along with location and a natural deep harbor.In [4], the design of a simulation model was presented and the model was used in a research project toinvestigate terminal structures for container handling on the planned extension of the Rotterdan portarea"Maasvlaktc 2".A new modeling technique was applied in the control of automated guidedvehicle (AGV) traffic at a container terminal in [5]. A simulation model for integrating quay transportand stacking policies on automated container terminals was put forward in [6]. In [7], on the basis ofacombination of three cycleoperations: quay-yard cycle, in-out gate cycle and front-rear yardtransposition cycle, a generic simulation modeling for the container terminals logistics operationssystem was proposed.Based on the review of literature, there are some papers researching relevant fields. However, asystem dynamics model (SDM), which analyzes port operation system from perspectives of timequality and profit, has not been involved.This paper considers the POs as a complex one,approaching it from the three perspectives:Time, profit, and quality, which are most basic and crucialfactors in port operation. It builds the SDM for POS consisting of a time submodel, a profit submodeland a profit submodel, and testifies the model by applying it on Lianyungang Port case simulated byVensim software. The three submodels are inter-related. Changes in any of them will definitely affecteachother.The purpose of this study is to provide a useful tool to help decision-maker to analyze the POS inorder to achieve the three management goals: guarantying service time, improving quality, andlowering cost and increasing profit.II.DEFINING“POSANDITSSUBMODELS”A.POSThe input-transformation-output process is characteristic of a wide variety of operating systemsFor port enterprises,the POs includes a set of factors related to port service hours, quality and profitsand can be in charge of planning, organizing, and controlling the input-transformation-output processand activities in port enterprises.Time, quality and profit are the key factors that could be employed to make a comprehensiveevaluation on port enterprise operation system. Therefore, in this paper, the POS is divided into timesubsystem (TS), quality subsystem (QS) and profit subsystem (PS), and the POS is defined as acombination of the three subsystems and of substance system and management system involved at theport.These three are inseparable and inter-dependent. Firstly, the TS includes the ship's berth-time andthe cargoes' port-time which are the main factors that ship-owners and cargo owners are most

study on Port of Singapore indicates a combination of resources can help create a sustainable advantage for the Port. Resources included supportive government policies, ample investment, and well thought out operations and information technology along with location and a natural deep harbor. In [4], the design of a simulation model was presented and the model was used in a research project to investigate terminal structures for container handling on the planned extension of the Rotterdan port area "Maasvlaktc 2". A new modeling technique was applied in the control of automated guided vehicle (AGV) traffic at a container terminal in [5]. A simulation model for integrating quay transport and stacking policies on automated container terminals was put forward in [6]. In [7], on the basis ofa combination of three cycleoperations: quay-yard cycle, in-out gate cycle and front-rear yard transposition cycle, a generic simulation modeling for the container terminals logistics operations system was proposed. Based on the review of literature, there are some papers researching relevant fields. However, a system dynamics model (SDM), which analyzes port operation system from perspectives of time, quality and profit, has not been involved. This paper considers the POS as a complex one, approaching it from the three perspectives：Time, profit, and quality, which are most basic and crucial factors in port operation. It builds the SDM for POS consisting of a time submodel, a profit submodel and a profit submodel, and testifies the model by applying it on Lianyungang Port case simulated by Vensim software. The three submodels are inter-related. Changes in any of them will definitely affect each other. The purpose of this study is to provide a useful tool to help decision-maker to analyze the POS in order to achieve the three management goals: guarantying service time, improving quality, and lowering cost and increasing profit. II. DEFINING “POSAND ITS SUBMODELS” A. POS The input-transformation-output process is characteristic of a wide variety of operating systems. For port enterprises, the POS includes a set of factors related to port service hours, quality and profits, and can be in charge of planning, organizing, and controlling the input-transformation-output process and activities in port enterprises. Time, quality and profit are the key factors that could be employed to make a comprehensive evaluation on port enterprise operation system. Therefore, in this paper, the POS is divided into time subsystem (TS), quality subsystem (QS) and profit subsystem (PS), and the POS is defined as a combination of the three subsystems and of substance system and management system involved at the port. These three are inseparable and inter-dependent. Firstly, the TS includes the ship’s berth-time and the cargoes’ port-time which are the main factors that ship-owners and cargo owners are most

concernedwith.Therefore,theyaretheimportantmeasuresoftheportservicelevel.Byreducingthetimeperiodsofthesetwo operationprocesses, customerswilldefinitelybemoresatisfied withportservices.Secondly,on the onehand, theQS can be affected byTS and it canalso be influenced byport safety quality as well as the quality of cargo loading-unloading and storage. On the other hand,the QS could also directly influence the PS. Poor service quality will not also increase quality cost butalso lead to decrease in port cargo freight, which means less profit. This will undoubtedly bring downthe Ps. Finally, the increase in port profits will bring about more port investments, more berths, moreequipment and larger storage coverage. By reducing ship's berth-time and the cargoes' port-time, thePS could have direct impact on theTS.B.TimeSubsystemThe TS includes the two different types of time:ship'sberthtime and thecargoesport-time.Aship's berth-time is referred to as: starting from the moment the ship berths till the ship departs fromthe pier after loading unloading cargoes if the ship berths alongside the pier directly after entering intotheport, startingfromthemomenttheshipberths attheanchoragearea orbuoytill the shipdepartsfrom the anchorage area or buoy after loading-unloading cargoes if the ship berths at the anchoragearea or buoy. A ship's berth-time time consists of the three parts: operational berth-time,non-operational berth-time and lay-off time caused by natural factors.The cargoes'port-time includes the cargoes'yard-time for storage and the cargoes'berth-time forstevedoring. Typically, the cargoes' port-time is influenced by the loading-unloading efficiency of theship, the warehouse capacity and the collecting and distributing volume by land.C.QualitySubsystemThe QS of port operation mainly refers to the security quality and service quality of a port.Elements used to evaluating the security quality of a port include the following four parts: 1) casualtyof workers, referring to the totality of people dying or injured in each accidents during the reportingperiod, 2) on-time delivery rate, referring to the ratio of the on-time delivery times of the port to thetotal delivery times during the reporting period, 3)damage rate of cargoes, referring to the ratio ofdamaged cargoes to the totality of cargoes during the reporting period, and 4) shortage rate of cargoes,referring to the ratio of freight shortage to the totality of cargoes during the reporting period.Moreover, factors used to evaluating the service quality of a port are referred to as theconvenience of a port's service, after-sales service provided by the port and psychological influenceuponcustoms.D.Profit SubsystemThe profit of port enterprises is the financial achievement obtained by enterprises' operation ofloading-unloading business, storing business, agency business and other businesses. In this paper, it isdefined as what the income of the main business subtracts the total cost of port operations. The total

concerned with. Therefore, they are the important measures of the port service level. By reducing the time periods of these two operation processes, customers will definitely be more satisfied with port services. Secondly, on the one hand, the QS can be affected by TS and it can also be influenced by port safety quality as well as the quality of cargo loading-unloading and storage. On the other hand, the QS could also directly influence the PS. Poor service quality will not also increase quality cost but also lead to decrease in port cargo freight, which means less profit. This will undoubtedly bring down the PS. Finally, the increase in port profits will bring about more port investments, more berths, more equipment and larger storage coverage. By reducing ship’s berth-time and the cargoes’ port-time, the PS could have direct impact on the TS. B. Time Subsystem The TS includes the two different types of time: ship’s berth time and the cargoes’ port-time. A ship’s berth-time is referred to as: starting from the moment the ship berths till the ship departs from the pier after loading unloading cargoes if the ship berths alongside the pier directly after entering into the port; starting from the moment the ship berths at the anchorage area or buoy till the ship departs from the anchorage area or buoy after loading-unloading cargoes if the ship berths at the anchorage area or buoy. A ship’s berth-time time consists of the three parts: operational berth-time, non-operational berth-time and lay-off time caused by natural factors. The cargoes’ port-time includes the cargoes’ yard-time for storage and the cargoes’ berth-time for stevedoring. Typically, the cargoes’ port-time is influenced by the loading-unloading efficiency of the ship, the warehouse capacity and the collecting and distributing volume by land. C. Quality Subsystem The QS of port operation mainly refers to the security quality and service quality of a port. Elements used to evaluating the security quality of a port include the following four parts: 1）casualty of workers, referring to the totality of people dying or injured in each accidents during the reporting period, 2）on-time delivery rate, referring to the ratio of the on-time delivery times of the port to the total delivery times during the reporting period, 3）damage rate of cargoes, referring to the ratio of damaged cargoes to the totality of cargoes during the reporting period, and 4）shortage rate of cargoes, referring to the ratio of freight shortage to the totality of cargoes during the reporting period. Moreover, factors used to evaluating the service quality of a port are referred to as the convenience of a port’s service, after-sales service provided by the port and psychological influence upon customs. D. Profit Subsystem The profit of port enterprises is the financial achievement obtained by enterprises’ operation of loading-unloading business, storing business, agency business and other businesses. In this paper, it is defined as what the income of the main business subtracts the total cost of port operations. The total

costsof portoperationsmainlyconsistof fixed costsand variablecosts.Fixedcosts includethesalaryemployee welfare funds, the insurance, depreciation fees and the rent charge of the loading-unloadingequipment. Variable costs include repairing fees, fuel costs, finance costs, management fees, laborprotectionfees and quality costs.Quality costs aretheexpenditureon securingthe satisfactoryqualityduring the operation of enterprises. They can be divided into quality guarantee costs and qualityfailure costs due to the poor quality.In terms of different kinds of income, the income of main business of port enterprises can beclassified into the following five types of source: 1) the loading and unloading income, 2) the storageincome, 3)the agency income,4)the port management income, and 5)other business income.III.BUILDINGASDMFORPOSTheSDMforPOScanbebuilt afterthethreedifferent submodelsofSDMareestablishedA.Sub model for TSThe ship's operational berth-time is affected by the three kinds of productivity and the three kindsof impact factors: (1) the daily productivity of stevedores (DPS) and its impact factor, (2) theproductivity per hour per loading-unloading equipment (PHE) and its impact factor, and 3) theproductivity per hour per vehicle (PHV) and its impact factor.The threekinds of the productivity canbe calculated by the processed statistics data. The three impact factors can be obtained by means ofmultiple linear regressions when the ship's operational berth-time is used as dependent variable andthe three impact factors as independent variables. The ship's non-operational berth-time is mainlyaffected by the berths resource allocation and its impact factor, and the equipment productivity and itsimpact factors. Similarly,the two impact factors can also be obtained by multiple linear regressions.Based on the above analysis, as one part of the sub model for TS, the SDM of the ship'sberth-time at the port is built and illustrated in Figure ITCETTESWDFPHETCVCOTFPHVSBTCBTFigurelSDMofSBTFactors shown in FigureI denote following things:SBT: the ship's berth-time.CBT:changes intheberth-timeoftheship

costs of port operations mainly consist of fixed costs and variable costs. Fixed costs include the salary, employee welfare funds, the insurance, depreciation fees and the rent charge of the loading-unloading equipment. Variable costs include repairing fees, fuel costs, finance costs, management fees, labor protection fees and quality costs. Quality costs are the expenditure on securing the satisfactory quality during the operation of enterprises. They can be divided into quality guarantee costs and quality failure costs due to the poor quality. In terms of different kinds of income, the income of main business of port enterprises can be classified into the following five types of source: 1) the loading and unloading income, 2) the storage income, 3) the agency income, 4) the port management income, and 5) other business income. III. BUILDINGA SDM FOR POS The SDM for POS can be built after the three different sub models of SDM are established. A. Sub model for TS The ship’s operational berth-time is affected by the three kinds of productivity and the three kinds of impact factors: (1) the daily productivity of stevedores (DPS) and its impact factor, (2) the productivity per hour per loading-unloading equipment (PHE) and its impact factor, and 3) the productivity per hour per vehicle (PHV) and its impact factor. The three kinds of the productivity can be calculated by the processed statistics data. The three impact factors can be obtained by means of multiple linear regressions when the ship’s operational berth-time is used as dependent variable and the three impact factors as independent variables. The ship’s non-operational berth-time is mainly affected by the berths resource allocation and its impact factor, and the equipment productivity and its impact factors. Similarly, the two impact factors can also be obtained by multiple linear regressions. Based on the above analysis, as one part of the sub model for TS, the SDM of the ship’s berth-time at the port is built and illustrated in Figure Ⅰ. Figure1 SDM of SBT Factors shown in Figure Ⅰdenote following things: SBT: the ship’s berth-time. CBT: changes in the berth-time of the ship

CNOT:changes in the non-operational berth-time of the shipCOT: changes in the operational berth-time of the shipFPHV:theimpactfactorofPHVFDPS: the impact factor of DPSFPHE: the impact factor of PHETP: the throughput of the port.TCV: the total tonnage of cargoes loaded-unloaded by vehicles.TTV: the total loading-unloading time of vehiclesOT: operation tons.NL:the numbers of labors.SWD:the stevedores'workdays.TTE: the total operating time of loading-unloading equipment.TCE: the total tonnage of cargoes handled by loading unloading equipmentBA: berths resource allocation.BC: causes of berths.FBC:the impactfactorofberthsOC: other causes.FOC: the impact factor of other causes.Moreover, the cargoes' port-time (CPT) is affected by: 1) SBT and its impact factor, 2) thewarehouse capacity (WC) and its impact factor, and 3) the collecting and distributing volume by land(VL) and its impact factor. Similarly, the capacity and the volume can calculated by the processedstatistics data, and the three impact factors can be obtained by multiple linear regressions. As the otherpart of the sub model for TS, the SDM of the cargoes' port-time is built and illustrated in Figure IISBTVRWFSBTVRDFWCWCFVLCCPTFigure2 SDM of CPTFactors shown in Figure II denote following thingsFWC:the impactfactor of warehouse capacityFSBT: the impact factor of SBT.FVL: the impact factor of the collecting and distributingvolume by land

CNOT: changes in the non-operational berth-time of the ship. COT: changes in the operational berth-time of the ship. FPHV: the impact factor of PHV. FDPS: the impact factor of DPS. FPHE: the impact factor of PHE. TP: the throughput of the port. TCV: the total tonnage of cargoes loaded-unloaded by vehicles. TTV: the total loading-unloading time of vehicles. OT: operation tons. NL: the numbers of labors. SWD: the stevedores’ work days. TTE: the total operating time of loading-unloading equipment. TCE: the total tonnage of cargoes handled by loading unloading equipment. BA: berths resource allocation. BC: causes of berths. FBC: the impact factor of berths. OC: other causes. FOC: the impact factor of other causes. Moreover, the cargoes’ port-time (CPT) is affected by: 1) SBT and its impact factor, 2) the warehouse capacity (WC) and its impact factor, and 3）the collecting and distributing volume by land (VL) and its impact factor. Similarly, the capacity and the volume can calculated by the processed statistics data, and the three impact factors can be obtained by multiple linear regressions. As the other part of the sub model for TS, the SDM of the cargoes’ port-time is built and illustrated in Figure Ⅱ. Figure2 SDM of CPT Factors shown in Figure Ⅱdenote following things: FWC: the impact factor of warehouse capacity. FSBT: the impact factor of SBT. FVL: the impact factor of the collecting and distributing volume by land

VRD: the collecting and distributing volumeby roadVRW:the collectingand distributingvolumeby railwayCCPT: changes in the average cargoes' port-time.B.Sub model for QSThe traditional concept of port quality only pays attention to security quality and freight quality ofa port, but the two factors are not the entire content about port quality at all. Besides the traditionalcriterion measuring the quality of a port, the paper adds more factors like customs'impression on theport, after-sales services, and the convenience of services in order to constitute comprehensivemeasurements system of port quality. The submodel for QS is built and illustrated in Figure III.RODCWRDSTTCADRPQCPIPCASSACPQIFCCPSEFC<CPTFigure3 Sub model for QSFactors shown in Figure IlIdenote following things:AD: accident damages.PC: prevention costs.AC: appraisal costs.IFC: internal failure costs.EFC: external failure costs.PQ: the port quality.RPQ: the rate of change in port qualityCw: casualty of workers.ROD: on-time delivery rate.RDS: the rate of freight damage and shortage.TTC: the total tonnage of cargoesCPI: the customs’psychological impressionASS: the after-sales service.CPS: the convenience of port servicesC.Sub model for PSBasedontheaboveanalysisofPS,thesubmodelforPSis

VRD: the collecting and distributing volumeby road. VRW: the collecting and distributing volume by railway. CCPT: changes in the average cargoes’ port-time. B. Sub model for QS The traditional concept of port quality only pays attention to security quality and freight quality of a port, but the two factors are not the entire content about port quality at all. Besides the traditional criterion measuring the quality of a port, the paper adds more factors like customs’ impression on the port, after-sales services, and the convenience of services in order to constitute comprehensive measurements system of port quality. The submodel for QS is built and illustrated in Figure Ⅲ. Figure3 Sub model for QS Factors shown in Figure Ⅲdenote following things: AD: accident damages. PC: prevention costs. AC: appraisal costs. IFC: internal failure costs. EFC: external failure costs. PQ: the port quality. RPQ: the rate of change in port quality. CW: casualty of workers. ROD: on-time delivery rate. RDS: the rate of freight damage and shortage. TTC: the total tonnage of cargoes. CPI: the customs’ psychological impression. ASS: the after-sales service. CPS: the convenience of port services. C. Sub model for PS Based on the above analysis of PS, the sub model for PS is

built and illustrated in Figure IVNEEPHENTDNYENLEFFLEKTPFBRFTNEEFINEESOFACTFI-SRCTPCTPFigure 4 Sub model for PSFactors shown in Figure IVdenote following things:LI: the labor insurance.SL: the salaryEW:employee welfaresDF:depreciationfeesIS:the insurance.RC:the rent chargeFC: fixed costs.LPF: the labor protection fee.EFL:the expenditure on fuel and lubricant.RF: repairing fees.MF: management fees.FCS: finance costs.VC: variable costs.QC: quality costs.TCO: the total costs of operations.NLE: the number of loading-unloading equipment.NYE:thenumber of yards equipment.NTV: the number of transportation vehicles.TNE: the total umber of equipmentFFLE: the impact factor of fuel and lubricant affected by the equipmentFFLT: the impact factor of fuel and lubricant affected by the throughput.FBR:theimpactfactorofberthrecourseallocation.FTNE:theimpactfactor of thetotal number ofequipment.NEE:thetotalnumberofemployees

built and illustrated in Figure Ⅳ. Figure 4 Sub model for PS Factors shown in Figure Ⅳdenote following things: LI: the labor insurance. SL: the salary. EW: employee welfares. DF: depreciation fees. IS: the insurance. RC: the rent charge. FC: fixed costs. LPF: the labor protection fee. EFL: the expenditure on fuel and lubricant. RF: repairing fees. MF: management fees. FCS: finance costs. VC: variable costs. QC: quality costs. TCO: the total costs of operations. NLE: the number of loading-unloading equipment. NYE: the number of yards equipment. NTV: the number of transportation vehicles. TNE: the total umber of equipment. FFLE: the impact factor of fuel and lubricant affected by the equipment. FFLT: the impact factor of fuel and lubricant affected by the throughput. FBR: the impact factor of berth recourse allocation. FTNE: the impact factor of the total number of equipment. NEE: the total number of employees

TFI: the total freight income.TP:total profitsCTP: Changes in the amount of total profits.OF: other factors.TNYFPHVICPTFigure5SDMforPOSD.ASDMofPOSAfter analyzing the three subsystems and the three corresponding submodels which are thecomponents of the system model, the SDM for POS is built and illustrated in Figure 5.IV.ACASESTUDYOne of the loading-unloading companies of Lianyungang Port is selected as a case study.Thisport's cargo throughput ranked the 9th among all the Chinese ports in 2008. It takes the processedstatistical data from 2001 to 2007 as input parameters for models, adopts the regression analysis toestablish related equations, and uses the system dynamics software Vensim PLE to conduct simulationupon models.ADataInputandProcessDue tothelimited space, the ship's operational berth-time istaken as an exampleto illustrate howto get related equation. After data fitting using Vensim tool on the DPS, PHE and PHV collected in thereal word,thefunction is found as:CBT=0.001×DPS-0.145×PHE-0.008×PHV+4.947(1)B.Sensitivity analysis of the modelFigure Villustrates that there are many variables in the SDM for POS and the change of eachvariable can all be conveyed through the model's feedback mechanism.Due to the limitation of spacethe paper only chooses the numbers of the port's equipment, a rather more important factor, to presentthe sensitivity analysis by increasing the number of port's main equipment by 10%. Seen in Figure VI,the sensitivity analysis of ship's berth-time shows that ship's berth-time is apparently reduced due to

TFI: the total freight income. TP: total profits. CTP: Changes in the amount of total profits. OF: other factors. Figure5 SDM for POS D. A SDM of POS After analyzing the three subsystems and the three corresponding submodels which are the components of the system model, the SDM for POS is built and illustrated in Figure 5. IV. A CASE STUDY One of the loading-unloading companies of Lianyungang Port is selected as a case study. This port’s cargo throughput ranked the 9th among all the Chinese ports in 2008. It takes the processed statistical data from 2001 to 2007 as input parameters for models, adopts the regression analysis to establish related equations, and uses the system dynamics software Vensim PLE to conduct simulation upon models. A. Data Input and Process Due to the limited space, the ship’s operational berth-time is taken as an example to illustrate how to get related equation. After data fitting using Vensim tool on the DPS, PHE and PHV collected in the real word, the function is found as: CBT=0.001×DPS-0.145×PHE-0.008×PHV+4.947 (1) B. Sensitivity analysis of the model FigureⅤillustrates that there are many variables in the SDM for POS and the change of each variable can all be conveyed through the model’s feedback mechanism. Due to the limitation of space, the paper only chooses the numbers of the port’s equipment, a rather more important factor, to present the sensitivity analysis by increasing the number of port’s main equipment by 10%. Seen in Figure Ⅵ, the sensitivity analysis of ship’s berth-time shows that ship’s berth-time is apparently reduced due to

the decrease in operational berth-time after the increase in loading- unloading equipment.as2beforetheincreasein1.75equipment1.5 aftertheincreaseinequipment1.2501030405060708091002YearFigure6.SensitivityAnalysisofSBTVarious elements in SDM for POS influence each other.The increase in equipment results in thereduction in fixed costs and the increase in variable costs. Figure Vllillustrates the change of the totalcosts of operation.The above sensitivity analysis shows that thenumber of mechanical configuration in theportenterprise is a little bit insufficient compared to its throughput. The appropriate increase in equipmentexerts the following influences upon the three subsystems: 1) It helps improve the TS and reduce theservice time for ships. 2) Influences upon the QS can be understood from two aspects: improving theservice level due to the reduction in the ships' berth-time; increasing the quality costs. 3) The TCO isnot greatly affected because the increase in quality costs is almost offset by the decrease in rentcharge.afterthe40increaseinequipment30beforetheincreaseinequipment2001020304050607080910YearFigure7 Sensitivity Analysis of TCOV.CONCLUSIONSThe POs is a quite complicated and dynamic system and involves a variety of factors, whichinteract with each other.This paper probes into internal structure and operation mechanism anddivides the complex system into TS, QS and PS; then, after the three sub models are built, the SDMfor POs is built via system dynamic; finally, through the case study and simulation, it seeks to findeffective ways in solving issues in the POS. The research indicates that the SDM for POS could beusedto:1)testthechainreactionincurredbychangingonecertainfactor,2)explorerelevantfactorsand the changes in the different sub models by changing one certain factor, 3) seek possible ways inadjusting and improving port operation level by comparing input-output and analyzing sensitivity.Besides, this paper also attempts to use statistical analysis and regression method to determine input

the decrease in operational berth-time after the increase in loading- unloading equipment. Figure 6. Sensitivity Analysis of SBT Various elements in SDM for POS influence each other. The increase in equipment results in the reduction in fixed costs and the increase in variable costs. Figure Ⅶillustrates the change of the total costs of operation. The above sensitivity analysis shows that the number of mechanical configuration in the port enterprise is a little bit insufficient compared to its throughput. The appropriate increase in equipment exerts the following influences upon the three subsystems: 1）It helps improve the TS and reduce the service time for ships. 2) Influences upon the QS can be understood from two aspects: improving the service level due to the reduction in the ships’ berth-time; increasing the quality costs. 3) The TCO is not greatly affected because the increase in quality costs is almost offset by the decrease in rent charge. Figure7 Sensitivity Analysis of TCO V. CONCLUSIONS The POS is a quite complicated and dynamic system and involves a variety of factors, which interact with each other. This paper probes into internal structure and operation mechanism and divides the complex system into TS, QS and PS; then, after the three sub models are built, the SDM for POS is built via system dynamic; finally, through the case study and simulation, it seeks to find effective ways in solving issues in the POS. The research indicates that the SDM for POS could be used to: 1) test the chain reaction incurred by changing one certain factor; 2) explore relevant factors and the changes in the different sub models by changing one certain factor; 3) seek possible ways in adjusting and improving port operation level by comparing input-output and analyzing sensitivity. Besides, this paper also attempts to use statistical analysis and regression method to determine input

parameters of the model, which increases the effectiveness and applicability of the SDMAs a generic model, it can be applicable to any port to analyze the POS and its managementpolicies,as well as providedecision-making supports for strategy and tacticsfor port enterprises

parameters of the model, which increases the effectiveness and applicability of the SDM. As a generic model, it can be applicable to any port to analyze the POS and its management policies, as well as provide decision-making supports for strategy and tactics for port enterprises