Risk management and inventory cost control are key issues in supply chain management. Based on an (r, Q) strategy, this paper formulates a multi-layer multi-period stochastic inventory problem as a robust multi-objective model. The novelty lies in the consideration of both perturbed variables and stochastic demand in the model. The goal is to minimize the expected cost and the risk measured by conditional value at risk (CVaR). To solve this model, we propose a hybrid Non-Dominated Genetic Algorithm-II (NSGA-II) where a polynomial time algorithm is designed to obtain the optimal CVaR for a given (r, Q). Moreover, a local search method is tailored for the NSGA-II to improve solutions. This hybrid algorithm can significantly increase the number of optimal solutions and decrease the inventory cost. Numerical results validate the effectiveness of our robust multi-objective model and the hybrid algorithm.
AcerbiATascheD (2002) On the coherence of expected shortfall. Journal of Banking and Finance26: 1487–1503.
2.
AdidaEPerakisG (2006) A robust optimization approach to dynamic pricing and inventory control with no backorders. Mathematical Programming107(1–2): 97–129.
3.
AhmedSCakmakUShapiroA (2007) Coherent risk measures in inventory problems. European Journal of Operational Research182(1): 226–238.
4.
AtznerPDelbaenFEberJM (2000) Coherent measures of risk. Mathematical Finance9: 203–228.
5.
AxsäterSZhangWF (1999) A joint replenishment policy for multi-echelon inventory control. International Journal of Production Economics59(1): 243–250.
6.
BrankeJ (1998) Creating robust solutions by means of evolutionary algorithms . In: Parallel Problem Solving from Nature, Proceedings PPSN V. Berlin Heidelberg: Springer, pp. 119–128.
7.
ChenFFedergruenA (2000) Mean-variance analysis of basic inventory models. Technical Manuscript, Columbia University.
8.
ChenYXuMZhangZG (2009) Technical note-a risk-averse newsvendor model under the CVaR criterion. Operations Research57(4): 1040–1044.
9.
DebKGuptaH (2006) Introducing robustness in multi-objective optimization. Evolutionary Computation14(4): 463–494.
10.
DebKPratapAAgarwalS. (2002) A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation6(2): 182–197.
FonsecaCMFlemingPJ (1993) Multiobjective genetic algorithms. In: IEE colloquium on genetic algorithms for control systems engineering (Digest No. 1993/130), London, UK, 28 May 1993. London: IEE.
13.
GumusTGuneriAF (2007) Multi-echelon inventory management in supply chains with uncertain demand and lead times: literature review from an operational research perspective. Proceedings of the Institution of Mechanical Engineers, Part B, Journal of Engineering Manufacture221(10): 1553–1570.
14.
HollandJH (1975) Adaptation in Natural and Artificial Systems. Englewood Cliffs: University of Michigan Press.
15.
JammerneggWKischkaP (2009) Risk preferences and robust inventory decisions. International Journal of Production Economics118(1): 269–274.
16.
KonakACoitDWSmithAE (2006) Multi-objective optimization using genetic algorithms: A tutorial. Reliability Engineering and System Safety91: 992–1007.
17.
LucianoEPeccatiLCifarelliDM (2003) VaR as a risk measure for multi-period static inventory models. International Journal of Production Economics81–82: 375–384.
18.
RockafellarRTUryasevS (2000) Optimization of conditional value-at-risk. Journal of Risk2:21–41.
19.
SchafferJD (1985) Multiple objective optimization with vector evaluated genetic algorithms. In: Proceedings of the 1st International Conference on Genetic Algorithms, 1985, pp. 93–100.
SrinivasNDebK (1994) Multiobjective optimization using nondominated sorting in genetic algorithms. Journal of Evolutionary Computing2(3): 221–248.
22.
TapieroCS (2005) Value at risk and inventory control. European Journal of Operational Research163: 769–775.
23.
TsutsuiSGhoshA (1997) Genetic algorithms with a robust solution searching scheme. IEEE Transactions on Evolutionary Computation1(3): 201–208.
24.
ZhangDXuHWuY (2009) Single and multi-period optimal inventory control models with risk-averse constraints. European Journal of Operational Research199(2): 420–434.
25.
ZhangYSongSZhangH. (2012) A hybrid genetic algorithm for two-stage multi-item inventory system with stochastic demand. Neural Computing and Applications21(6): 1087–1098.
26.
ZitzlerEThieleL (1999) Multiobjective evolutionary algorithms: a comparative case study and the strength Pareto approach. IEEE Transactions on Evolutionary Computation3(4): 257–71.
27.
ZitzlerELaumannsMThieleL (2001) SPEA2: improving the strength Pareto evolutionary algorithm. Swiss Federal Institute Technology, Zurich, Switzerland.
