University of Exeter
Browse

Global dual sourcing: ARMA(1,1) market demand and its decomposition

Download (1.15 MB)
conference contribution
posted on 2025-08-02, 11:42 authored by T Hosoda, SM Disney, G Gaalman
We consider the case of reshoring, where a global firm takes back a portion of its low-cost offshore supply to be produced in nearshore factory in order to establish a dual sourcing supply chain equipped with both low cost and responsiveness. We first establish the performance benchmark of a single (nearshore or offshore) supplier model. In the dual sourcing setting, a firm decomposes the first order auto-regressive moving average, ARMA(1,1), market demand process into two parts: one for the nearshore source and one for the offshore source. Order-up-to policies determine the order quantities for both sources. We show how to reduce inventory costs in the dual-sourcing case to a level identical to the near-shore single-source case. Furthermore, if certain conditions are met, the nearshore manufacturing cost reduces in the offshore lead-time. This suggests low-cost and low-emission transport modes should be utilized (slow steaming vessels which are both low cost and environmentally friendly, but may endure longer offshore lead-times come to mind), breaking the trade-off between economic and environmental performance.

History

Rights

© 2024 International Society for Inventory Research

Notes

This is the author accepted manuscript.

Publisher

International Society for Inventory Research

Name of conference

23rd International Working Seminar on Production Economics

Location

Innsbruck, Austria

Version

  • Accepted Manuscript

Language

en

FCD date

2024-03-06T09:08:15Z

FOA date

2024-03-06T10:33:24Z

Citation

23rd International Working Seminar on Production Economics, 14 - 18 February 2024, Innsbruck, Austria, pp. 50 - 65

Department

  • Management

Usage metrics

    University of Exeter

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC