A multi-criteria approach to robust outsourcing decision-making in stochastic manufacturing systems

• Robust DEA-based approach for multi-criteria decision-making developed.
• Partial views of DEA envelope curves provide instructive decision support.
• Enhanced aggregate planning approach for stochastic environments developed.
• Demand variability drives outsourcing volumes and reduces internal batch sizes.
• Higher setup variability increases insourcing volumes and average batch sizes.

Manufacturing outsourcing is a key industry trend towards greater operations effectiveness and is related to the discussion of strategic core competencies. We study the issue of contract manufacturing at the strategic–tactical level aiming for robust decisions to accommodate stochastic manufacturing environments and immanent uncertainty of planning parameters. The topic is approached from a multi-criteria decision-making perspective, since service, cost, quality, and more long-term value-related aspects need to be considered to arrive at well-balanced decisions. Our contribution is twofold: first, we develop a scenario-based non-parametric ranking approach to determine beneficial outsourcing options at the strategic level. The ranking method uses both model-based Key Performance Indicators (KPIs), which are obtained from a tactical planning model, and non-model-based KPIs that are derived in an independent assessment from multiple stakeholders. Second, we provide an enhanced aggregate planning approach at the tactical level in order to evaluate the performance implications of the strategic outsourcing decisions which in turn serve as the model-based KPIs for the ranking method. A queuing network-based approach is incorporated in the aggregate planning model to anticipate the stochastic behavior of manufacturing systems. An industry-derived case example with distinct outsourcing options is used to highlight the benefits of the approach and to investigate tactical trade-offs when coordinating internal and external manufacturing decisions.