Sourcing strategy and material control: An integrated model

This dissertation presents an analytical model that can help derive optimal sourcing strategy alternatives for firms that produce various finished products by assembling them from components which are purchased from external suppliers. A hierarchical, stochastic dynamic programming formulation has been developed to evaluate the impact of given sourcing strategy alternatives on the firm\u27s total operating costs and delivery service levels to customer in an environment of uncertain demand. These alternatives specify the length of contracts, purchase price schedules and delivery lead time schedules for all components. Our analysis can be used to evaluate the tradeoffs between the flexibility of short term, spot market mediated contracts, and the benefits of long term cooperative contracts between buyers and suppliers. Our formulation integrates strategic decisions with respect to selection of supplier contracts, and tactical decisions relating to material control and component allocation. A survey of eight leading manufacturing firms was conducted as part of this research to validate our modeling assumptions, and test the results of our analysis. In order to understand the impact of strategic sourcing decisions of the firm\u27s tactical decisions, a sub problem formulation of the firm\u27s assembly system has been developed. Consistent with the trend in many industries, commonality of components across different finished products is assumed. Resupply lead times for the different components may be different depending upon the suppliers chosen. The firm is assumed to operate in an environment of demand uncertainty for finished products, though the demands may be correlated within any period. We have studied the tradeoffs between total inventory and shortage costs over all components, and the resulting finished product delivery service metrics such as delays due to component shortages, fill rates and order completion rates. The results are used to determine the optimal order up to levels for components so that target levels of delivery service may be provided. Our results can also be used to study the impact of system parameters such as lead times, holding cost rates and target service levels on total costs