Use this URL to cite or link to this record in EThOS:
Title: Portfolio analysis in supply chain management of a chemicals complex in Thailand
Author: Suwanapal, Pasant
ISNI:       0000 0004 2700 137X
Awarding Body: Imperial College London
Current Institution: Imperial College London
Date of Award: 2010
Availability of Full Text:
Access from EThOS:
Access from Institution:
There is a considerable amount of research literature available for the optimisation of supply chain management of the chemical process industry. The context of supply chain considered in this thesis is the supply chain inside the chemical complex which is the conversion of raw materials into intermediate chemicals and finished chemical products through different chemical processes. Much of the research in the area of planning and scheduling for the process sector has been focused on optimising an individual chemical process within a larger network of a chemicals complex. The objective of this thesis is to develop a multi-objective, multi-period stochastic capacity planning model as a quantitative tool in determining an optimum investment strategy while considering sustainability for an integrated multi-process chemicals complex under future demand uncertainty using the development of inorganic chemicals complex at Bamnet Narong, Thailand as the main case study. Within this thesis, a number of discrete models were developed in phases towards the completion of the final multi-objective optimisation model. The models were formulated as mixed-integer linear programming (MILP) models. The first phase was the development of a multi-period capacity planning optimisation model with a deterministic demand. The model was able to provide an optimal capacity planning strategy for the chemicals complex at Bamnet Narong, Thailand. The numerical results show that based on the model assumptions, all the proposed chemical process plants to be developed in the chemicals complex are financially viable when the planning horizon is more than 8 years. The second phase was to build a multi-period stochastic capacity planning optimisation model under demand uncertainty. A three-stage stochastic programming approach was incorporated into the deterministic model developed in the first phase to capture the uncertainty in demand of different chemical products throughout the planning horizon. The expected net present value (eNPV) was used as the performance measure. The results show that the model is highly demand driven. The third phase was to provide an alternative demand forecasting method for capacity planning problem under demand uncertainty. In the real-world, the annual increases in demand will not be constant. A statistical analysis method named “Bootstrapping” was used as a demand generator for the optimisation model. The method uses historical data to create values for the future demands. Numerical results show that the bootstrap demand forecasting method provides a more optimistic solution. The fourth phase was to incorporate financial risk analysis as constraints to the previously developed multi-period three-stage stochastic capacity planning optimisation model. The risks associated with the different demand forecasting methods were analysed. The financial risk measures considered in this phase were the expected downside risk (EDR) and the mean absolute deviation (MAD). Furthermore, as the eNPV has been used as the usual financial performance measure, a decisionmaking method, named “Minimax Regret” was applied as part of the objective function to provide an alternative performance measure to the developed models. Minimax Regret is one kind of decision-making theory, which involves minimisation of the difference between the perfect information case and the robust case. The results show that the capacity planning strategies for both cases are identical Finally, the last phase was the development of a multi-objective, multi-period three stage stochastic capacity planning model aiming towards sustainability. Multiobjective optimisation allows the investment criteria to be traded off against an environmental impact measure. The model values the environmental factor as one of the objectives for the optimisation instead of this only being a regulatory constraint. The expected carbon dioxide emissions was used as the environmental impact indicator. Both direct and indirect emissions of each chemical process in the chemicals complex were considered. From the results, the decision-makers will be able to decide the most appropriate strategy for the capacity planning of the chemicals complex.
Supervisor: Shah, Nilay Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral