This dissertation presents the application of supply chain optimization to chemical industries like refinery industry in order to increase the total profit and decrease the costs in supply chains. Supply chains are divided into three categories such as ‘source’, ‘production’ and ‘distribution’, and supply chain optimization problems for each category are treated.
First, ‘Production’ is addressed. A scheduling of actual size refinery processes considering environmental impacts with multiobjective optimization is proposed. The effective scheduling of the objective of which is to maximize the total profit is needed for large-scale industry. In addition, companies cannot avoid making an effort to reduce environmental impacts. However, the two objectives are conflicting. In this case, the best way is to obtain Pareto optimal solutions by multiobjective optimization. In this thesis, a mixed-integer programming model is developed using the data from actual size refinery processes. ‘Critical Surface-Time 95’ that is one of the methods for environmental impacts valuation is used. ε-constraint method is used to solve the multiobjective optimization problem. Decision makers can find out the correlation between profitability and environment from the Pareto optimal solutions.
Second, ‘Distribution’ is addressed. A distribution network optimization considering continuous/discontinuous piecewise linear cost function is proposed. There are previous researches on distribution network optimization, the calculations of transportation cost and installation cost for facilities such as distribution centers and warehouses are too simple to reflect the real situations. In this thesis, the optimal distribution network is obtained with realistic transportation and installation calculation using ‘economy of scale’ concept.
Third, ‘Source’ is addressed. An environmental purchasing strategy for green SCM is proposed. Environmental purchasing is the first step to environmental products. It ...