In order to achieve stable and sustainable systems for recycling post-consumer goods, frequently it is necessary to concentrate the flows from many collection points of suppliers to meet the volume requirements for the recycler. The collection network must be grown over time to maximize the collection volume while keeping costs as low as possible. This paper addresses a complex and interconnected set of decisions that guide the investment in recruiting effort. Posed as a stochastic dynamic programming problem, the recruitment model captures the decisions for the processor who is responsible for recruiting material sources to the network. A key feature of the model is the behavior of the collector, whose willingness to join the network is modeled as a Markov process. An exact method and two heuristics are developed to solve this problem, then their performance is compared in solving practically sized problems.
Citation
Working paper, Industrial and Systems Engineering, Georgia Institute of Technology, 1/2007
Article
View Recruiting Suppliers for Reverse Production Systems: an MDP Heuristics Approach