Multi-level reverse logistics network design under uncertainty

Localising facilities and assigning product flows in a reverse logistics environment is a crucial but difficult strategic management decision, certainly when value decay plays an important part. Despite numerous publications regarding closed-loop supply chain design, very few addressed the impact of lead times and the high level of uncertainty in reverse processes. In this article, a single product reverse logistics network design problem with multiple layers and multiple routings is considered. To this end, a new advanced strategic planning model with integrated queueing relationships is built that explicitly takes into account stochastic delays due to various processes like collection, production and transportation, as well as disturbances due to various sources of variability like uncertain supply, uncertain process times, unknown quality, breakdowns, etc. Their impact is measured by transforming these delays into work-in-process, which affects profit through inventory costs. This innovative modeling approach is difficult to solve because of both combinatorial and nonlinear continuous relationships. The differential evolution algorithm with an enhanced constraint handling method is proposed as an appropriate heuristic to solve this model close to optimality. A number of scenarios for a realistic case illustrate the power of this optimisation tool.

[1]  Feng-Sheng Wang,et al.  A hybrid method of evolutionary algorithms for mixed-integer nonlinear optimization problems , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[2]  S. Seuring,et al.  End-of-life returns of long-lived products from end customer—insights from an ideally set up closed-loop supply chain , 2007 .

[3]  J. A. Buzacott,et al.  Open queueing network models of dynamic job shops , 1981 .

[4]  M. J. D. Powell,et al.  Algorithms for nonlinear constraints that use lagrangian functions , 1978, Math. Program..

[5]  Nico Vandaele,et al.  Clips: a capacity and lead time integrated procedure for scheduling , 1998 .

[6]  Sergio Rubio,et al.  Characteristics of the research on reverse logistics (1995–2005) , 2008 .

[7]  W. Whitt,et al.  The Queueing Network Analyzer , 1983, The Bell System Technical Journal.

[8]  Ivan Zelinka,et al.  Mechanical engineering design optimization by differential evolution , 1999 .

[9]  Nico Vandaele,et al.  Reverse logistics network design with stochastic lead times , 2007, Comput. Oper. Res..

[10]  B. Babu,et al.  A DIFFERENTIAL EVOLUTION APPROACH FOR GLOBAL OPTIMIZATION OF MINLP PROBLEMS , 2002 .

[11]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[12]  Nico Vandaele,et al.  Optimal grouping for a nuclear magnetic resonance scanner by means of an open queueing model , 2003, Eur. J. Oper. Res..

[13]  Zbigniew Michalewicz,et al.  Evolutionary Algorithms for Constrained Parameter Optimization Problems , 1996, Evolutionary Computation.

[14]  Jo van Nunen,et al.  Special Section: Closed-Loop Supply Chains: Practice and Potential: Integrating Closed-Loop Supply Chains and Spare-Parts Management at IBM , 2002, Interfaces.

[15]  J. Lampinen.  Multi-Constrained Nonlinear Optimization by the Differential Evolution Algorithm , 2002 .

[16]  Nico Vandaele,et al.  Optimal grouping at a nuclear magnetic resonance scanner , 2001 .

[17]  Rainer Leisten.  Production and Operations Management 2 , 2011 .

[18]  John A. Buzacott,et al.  Stochastic models of manufacturing systems , 1993 .

[19]  Ward Whitt,et al.  APPROXIMATIONS FOR THE GI/G/m QUEUE , 1993 .

[20]  Gilvan C. Souza,et al.  Time Value of Commercial Product Returns , 2006, Manag. Sci..