MIT ESD 71 Application Portfolio Brian R. Ramos Berthing Pain Prevention: Flexibility in the Design of a Marine Dock for Bulk Liquid Chemicals Loading Abstract: In this paper, we analyze the value of flexibility in the design of a marine dock for the loading of bulk chemicals from a riverside petrochemical manufacturing facility in the Gulf Coast region of the United States. After describing the general operating context and constraints, we investigate the output of a stochastic simulation model, comparing the expected financial impact of design approaches with and without built-in expansion flexibility. Additionally, we provide accompanying sensitivity analysis for the various uncertainties. The outcomes revealed through Monte Carlo simulation show that flexibility in the original design, combined with an option for later expansion, delivers the optimal expected net present value. December 7, 2010MIT ESD 71 Application Portfolio Brian R. Ramos TABLE OF CONTENTS INTRODUCTION Background Methodology Data Sources, Uncertainties and Parameters ANALYSIS Inflexible, Low Capacity Inflexible, High Capacity Flexible, Single Stage Flexible, Multi Stage Comparison CONCLUSION WORKS CITED Appendix A Appendix B MIT ESD 71 Application Portfolio Brian R. Ramos INTRODUCTIONBackground Though petrochemical manufacturing operations utilize various outbound transport modes—pipelines, tanker trucks, tanker rail cars, etc.—the primary transportation “workhorse of chemical logistics” is bulk marine shipping via ocean-going vessels and river barges (2118 Li et al 2010, 1365 Arons et al. 2006). Because of this reliance on water transport, bulk marine loading facilities can become a very expensive bottle-neck for a chemical manufacturer: vessel holdovers fees (demurrage) are in the tens of thousands of dollars per day while the opportunity costs of lost production capacity run even higher (1268 Jetlund 2004). Unfortunately, the high capital and operating costs imposed by these structures prohibits the profitable maintenance of considerable excess capacity. Typical marine loading facilities for petrochemicals consist of four primary components: berths/moorings, loading arms, pipelines, and a platform (e.g. piles and decking). As seen in Figures 1 and 2, a single platform typically hosts multiple pipelines, berths/moorings and loading arms, though the number varies by deployment. In our analysis, we attempt to exploit this componentization in order to gain the flexibility to incrementally expand marine loading capacity. Figure 1 Figure 2MIT ESD 71 Application Portfolio Brian R. Ramos MethodologyThe deterministic engineering design approach to such a capital construction project will seek the optimal fixed capacity which maximizes the value of the loading facility within the chosen timeframe. We use a stochastic simulation model to compare this deterministic approach to two designs which incorporate flexible expansion options. Our model incorporates several constant variables in addition to the stochastic variables (see Table 1). These variables are described in further detail in the following section. We demonstrate two approaches to implementing expansion flexibility with our model. Both approaches rely on additional initial capital investment in Period 0. This additional investment represents the cost of installing two components with excess capacity—the platform and pipeline. This staging in Period 0 allows for the addition of incremental loading arms and berth/mooring systems at a later date, based on increases in product demand. The first flexibility scenario demonstrates the option of making a single additional investment (i.e. adding two additional loading arms and equivalent berthing/mooring capacity at one time) in the loading system. The second flexibility scenario allows for multiple, smaller investments (i.e. adding the two additional loading arms and equivalent berthing/mooring capacity one at a time). The decision to execute the expansion options, or not, is made programmatically during each period based on the level of demand compared to the level of capacity. Our comparative analysis includes the expected NPV and discounted capital expenditure of each scenario. As an appendix (B) we provide a sensitivity analysis of the variables and the distributions of select stochastic variables. Microsoft Excel and Palisade @Risk provide the environment for our modeling
View Full Document