SGER: Forecast-Robust Capacity Acquisition and Subcontracting Methods

点击翻译按钮获取中文摘要

This Small Grant for Exploratory Research (SGER) provides funding for the development of methodologies for making forecast-robust capacity acquisition and subcontracting decisions in production planning. The strategic tradeoffs between acquiring capacity over a finite planning horizon, holding inventories and subcontracting orders to third party vendors will be examined for production planning problems. The research project has three main objectives: (1) investigating the structure of optimal solutions for capacity acquisition and subcontracting problems, (2) studying the robustness of the optimal and approximate solutions under demand forecast errors, (3) developing efficient methods that will produce solutions that are relatively insensitive to errors in the demand forecasts. Different types of cost structures that are appropriate for capacity acquisition, production, subcontracting and inventory holding decisions will be delineated. The computational complexity of the developed methods will be investigated for various subclasses. Extensive computational experiments will be conducted to test the effectiveness of the analytical results in improving the capacity acquisition and subcontracting decisions. With the insights gained from this research, the downstream impacts of forecast-robust capacity acquisition and subcontracting decisions will be investigated in multi-echelon vendor-customer supply chains. Since the capacity and subcontracting decisions of a firm delivering a final product have a trickling effect on the demand of its vendors, research in coordinated decision making throughout the supply chain of the product is essential in reducing the total manufacturing cost of the product. This research has the potential of providing analytical insights to the strategic tradeoffs in forecast-robust capacity acquisition andsubcontracting decisions, especially in service industry and high-tech industries with short product life span.