Article: Aggregate production planning problem for a hybrid stochastic system under partially observed state variables Journal: International Journal of Business Performance and Supply Chain Modelling (IJBPSCM) 2023 Vol.14 No.4 pp.421 - 449 Abstract: This paper considers the complexities of multi-period aggregate production planning in hybrid systems, merging manufacturing and remanufacturing processes. It addresses the challenges of partially observable inventory and stochastic constraints. Key contributions include applying the certainty-equivalence principle and Kalman estimator-derived statistics, transforming the stochastic problem into a deterministic one for a more straightforward resolution. Additionally, it introduces an open-loop strategy for periodic updates to the deterministic solution, providing a pragmatic approach to the stochastic issue. The research models a chance-constrained linear-quadratic Gaussian (LQG) problem, dealing with imperfect information in a make-to-stock scenario. Findings reveal that the open-loop updating method enhances managerial decision-making over non-updating strategies. A sensitivity analysis underscores the significant impact of manufacturing and remanufacturing costs on solutions. The study highlights that quasi-optimal solutions are useful for decision-makers, maintaining crucial aspects of the stochastic problem. This paper contributes important insights into aggregate production planning for hybrid systems. Inderscience Publishers - linking academia, business and industry through research

Title: Aggregate production planning problem for a hybrid stochastic system under partially observed state variables

Authors: Oscar Salviano; Frederic Andres

Addresses: School of Business and Economics, Pontifical Catholic University of Campinas, Campinas, SP, Brazil ' Digital Content and Media Sciences Research Division, National Institute of Informatics, Tokyo, Japan

Abstract: This paper considers the complexities of multi-period aggregate production planning in hybrid systems, merging manufacturing and remanufacturing processes. It addresses the challenges of partially observable inventory and stochastic constraints. Key contributions include applying the certainty-equivalence principle and Kalman estimator-derived statistics, transforming the stochastic problem into a deterministic one for a more straightforward resolution. Additionally, it introduces an open-loop strategy for periodic updates to the deterministic solution, providing a pragmatic approach to the stochastic issue. The research models a chance-constrained linear-quadratic Gaussian (LQG) problem, dealing with imperfect information in a make-to-stock scenario. Findings reveal that the open-loop updating method enhances managerial decision-making over non-updating strategies. A sensitivity analysis underscores the significant impact of manufacturing and remanufacturing costs on solutions. The study highlights that quasi-optimal solutions are useful for decision-makers, maintaining crucial aspects of the stochastic problem. This paper contributes important insights into aggregate production planning for hybrid systems.

Keywords: aggregate production planning; stochastic problem; Kalman filter; manufacturing/remanufacturing process; sub-optimal approach.

DOI: 10.1504/IJBPSCM.2023.136624

International Journal of Business Performance and Supply Chain Modelling, 2023 Vol.14 No.4, pp.421 - 449

Accepted: 18 Oct 2023
Published online: 08 Feb 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Aggregate production planning problem for a hybrid stochastic system under partially observed state variables

Keep up-to-date