Technical Memorandum Number 819 Centralization vs . Competition in Subcontracting Operations

Consider a set of manufacturers all of which can outsource part of their workload to a third-party in first-come-first-serve order, each with the objective of completing his workload the soonest possible. The resulting manufacturer strategies are not optimal for the system and especially, they do not necessarily result to maximum utilization of the third-party capacity. For this reason, the third-party finds an optimal centralized schedule and devises a savings sharing scheme that makes it more profitable for all manufacturers to follow the centralized schedule rather than the original. In this article we develop optimal centralized strategies and incentive schemes. Then, we perform a computational experiment that quantifies the value of coordination in this production chain first against the original schedule and then against Nash equilibrium schedules under complete or incomplete information. We found that more information benefits dramatically the third-party and the players individually but that they dissipate quickly. Also, we find that competition with complete information under-utilizes third-party capacity by about 7.5% and severely for incomplete information. In contrast, randomly generated schedules without information sharing underutilize the third-party by about 6%. However, Nash (FCFS) schedules provide very strong (weaker) incentives for coordination because our allocation scheme allocates 70% (50%)of the savings to the losers who comprise 53% (42%) of the manufacturers. Our findings lead to policy guidelines that facilitate centralization by allotting a large portion of the savings to the losers. ∗Weatherhead School of Management, Department. of Operations, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106-7235, gxv5@case.edu †Decision Sciences Department, Lundquist College of Business, 1208 University of Oregon, Eugene, OR 974031208, tolga@uoregon.edu