Data driven robust optimization for handling uncertainty in supply chain planning models

Gumte, K.M. and Mitra, K. et al. (2021) Data driven robust optimization for handling uncertainty in supply chain planning models. Chemical Engineering Science, 246. p. 116889. ISSN 00092509

Full text not available from this repository. (Request a copy)

Abstract

While addressing supply chain planning under uncertainty, Robust Optimization (RO) is regarded as an efficient and tractable method. As RO involves calculation of several statistical moments or maximum / minimum values involving the objective functions under realizations of these uncertain parameters, the accuracy of this method significantly depends on the efficient techniques to sample the uncertainty parameter space with limited amount of data. Conventional sampling techniques, e.g. box/budget/ellipsoidal, work by sampling the uncertain parameter space inefficiently, often leading to inaccuracies in such estimations. This paper proposes a methodology to amalgamate machine learning and data analytics with RO, thereby making it data-driven. A novel neuro fuzzy clustering mechanism is implemented to cluster the uncertain space such that the exact regions of uncertainty are optimally identified. Subsequently, local density based boundary point detection and Delaunay triangulation based boundary construction enable intelligent Sobol based sampling to sample the uncertain parameter space more accurately. The proposed technique is utilized to explore the merits of RO towards addressing the uncertainty issues of product demand, machine uptime and production cost associated with a multiproduct, and multisite supply chain planning model. The uncertainty in supply chain model is thoroughly analysed by carefully constructing examples and its case studies leading to large scale mixed integer linear and nonlinear programming problems which were efficiently solved in GAMS® framework. Demonstration of efficacy of the proposed method over the box, budget and ellipsoidal sampling method through comprehensive analysis adds to other highlights of the current work.

[error in script]

IITH Creators:

IITH Creators	ORCiD
Mitra, Kishalay	http://orcid.org/0000-0001-5660-6878

Item Type:

Article

Uncontrolled Keywords:

Chain management, Data driven robust optimization, Multilayered perceptrons, Neuro-fuzzy clustering, Parameter spaces, Robust optimization, Supply chain planning, Uncertain parameters, Uncertainty, Uncertainty modeling

Subjects:

Chemical Engineering

Divisions:

Department of Chemical Engineering

Depositing User:

Mrs Haseena VKKM

Date Deposited:

09 Dec 2021 05:54

Last Modified:

04 Mar 2022 04:56

URI:

http://raiithold.iith.ac.in/id/eprint/9030