A novel heuristic approach for planning decentralised supply chain under uncertainties

Abstract

Coordinating various activities among company members facing real-life uncertainties or disruptions is a great issue of concern in today’s business world. With this background, a multi-stage decentralised supply chain (SC) network is studied in this paper, where demand uncertainties are considered in each stage of the chain. We consider a serial SC network with single independent entities (manufacturer – distributor – retailer) in each level under restricted information sharing characteristics. The increased variability of uncertain demand through upward sections of the chain is studied. A two-phase planning model is proposed to coordinate the independent members with random customer demand. We develop a scenario-based stochastic optimisation approach where a probability is assigned for each scenario. A rolling horizon-based dynamic updating approach is proposed to update the model results for the current period as uncertainties are revealed. We develop a rule-based solution heuristic and conduct numerical analyses to validate the model. Our results are compared with two approaches – deterministic with mean demand and centralised structure with multiple scenarios. The comparative result shows that our model provides better feasible results with fewer shortage costs. Also, sensitivity analyses are performed on important parameters to observe their effect on the model.

KEYWORDS:

• Decentralised supply chain
• Demand uncertainty
• Heuristic
• Scenario-based analysis

Data availability statement

All data are included inside the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Marjia Haque

Marjia Haque is a Casual Academic in the School of Engineering and IT (SEIT) at UNSW Canberra, Australia. Her research interests include supply chain management, operations management, applied operations research and decision analytics.

Sanjoy Kumar Paul

Sanjoy Kumar Paul is an Associate Professor at the UTS Business School, University of Technology Sydney, Sydney, Australia. His research interests include sustainable and resilient supply chains, applied operations research, modelling and simulation, and intelligent decision-making.

Ruhul Sarker

Ruhul Sarker is a Professor in the School of Engineering and IT (SEIT) at UNSW Canberra, Australia. His broad research interests include decision analytics, CI / evolutionary computation, operations research, and applied optimisation.

Daryl Essam

Daryl Essam a Senior Lecturer in the School of Engineering and IT (SEIT) at UNSW Canberra, Australia. His research interests include genetic algorithms, with a focus on both evolutionary optimisation and large-scale problems.