We extend the linear programming (LP) model of deterministic supply-chain planning to take demand uncertainty and cash flows into account for the medium term. The resulting stochastic LP model is similar to that of Asset-Liability Management (ALM), for which the literature using stochastic programming is extensive. As such, we survey various modeling and solution choices developed in the ALM literature and discuss their applicability to supply chain planning. This survey can be a basis for making modeling/solution choices in research and in practice to manage the risks pertaining to unmet demand, excess inventory and cash liquidity when demand is uncertain.

https://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID1409647_code642908.pdf?abstractid=910579&mirid=3

Modeling Supply Chain Planning Under Demand Uncertainty Using Stochastic Programming: A Survey Motivated by Asset-Liability Management

Modeling supply-chain planning under demand uncertainty using stochastic programming: A survey motivated by asset–liability management

[1] Effective process control is crucial in implementing remediation actions for petroleum-contaminated sites. However, in dealing with in situ bioremediation practices, difficulties exist in incorporating numerical simulation models that are needed for process forecasting within real-time non-linear optimization frameworks that are critical for supporting the process control. With such difficulties, it is desired that a statistical relationship between remediation system performance and operating condition be established. Nevertheless, in the remediation systems, many variables can be either continuous or discrete, and the relations among them can be either linear or non-linear. These lead to complexities in the related multivaraite analyses. In this study, a forecasting system has been developed for supporting remediation design and process control based on techniques of NAPL-biodegradation (non-aqueous phase liquid biodegradation) simulation and stepwise-cluster analysis (SCA). The results indicate that the developed system is effective in forecasting the effects of multiple cleanup actions under various conditions. The predicted benzene concentrations have acceptable error levels compared with the outputs of numerical simulation. An optimization model for obtaining optimum operating conditions is then proposed to illustrate how the SCA method can be used for supporting optimization of bioremediation operations. A unique contribution of this research is the development of a multivariate inference system associated with simulation and optimization efforts for tackling the complex in situ bioremediation practices.

https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2005WR004006

Development of a forecasting system for supporting remediation design and process control based on NAPL-biodegradation simulation and stepwise-cluster analysis

https://tel.archives-ouvertes.fr/tel-01197082/document

Reducing myopic behavior in FMS control : a semi-heterarchical simulation-optimization approach

Purpose




This paper aims to propose a theoretical decision support framework, which integrates artificial intelligence (AI), discrete-event simulation (DES) and database management technologies so as to determine the steady state flow of items (e.g. fixtures, jigs, tools, etc.) in manufacturing.




Design/methodology/approach




The existing literature was carefully reviewed to address the state of the arts in decision support systems (DSS), the shortcomings of pure simulation-based and pure AI-based DSS. A conceptual example is illustrated to show the integrated application of AI, simulation and database components of the proposed DSS framework.




Findings




Recent DSS studies have revealed the limitations of pure simulation-based and pure AI-based DSS. A new DSS framework is required in manufacturing to address these limitations, taking into account the problems of flowing items.




Research limitations/implications




The theoretical DSS framework is proposed using simple rules and equations. This implies that it is not complex for software development and implementation. Practical data are not presented in this paper. A real DSS will be developed using the proposed theoretical framework and realistic results will be presented in the near future.




Originality/value




The proposed theoretical framework reveals how the integrated components of DSS can work together in manufacturing in order to determine the stable flow of items in a specific production period. Especially, the integrated performance of case-based reasoning (CBR) and DES is conceptually illustrated.

Decision support systems in manufacturing: a survey and future trends

Integration of simulation and optimization for solving complex decision making problems

This paper presents the ongoing development of a modeling methodology and a tool (the so-called simulation integrated system with modeling and optimization (SISMO) solver) that permits manufacturing systems to be both simulated and optimized according to several improvement strategies. We point out that the different steps of modeling, simulating, and optimizing uses the same integrated formalism and environment. A major point of this methodology and tool is the original decision-making mechanism over a hierarchy of complex discrete systems that model the real world. For the SISMO platform to be validated, we applied it to an actual highly constrained discrete-continuous scheduling problem. This study on a real-life problem has systematically led to improvements.

V. Bachelet

Papers

Integration of simulation and optimization for solving complex decision making problems

Integrating simulation and optimization of manufacturing systems