Supply chain risk management (SCRM) encompasses a wide variety of strategies aiming to identify, assess, mitigate and monitor unexpected events or conditions which might have an impact, mostly adverse, on any part of a supply chain. SCRM strategies often depend on rapid and adaptive decision-making based on potentially large, multidimensional data sources. These characteristics make SCRM a suitable application area for artificial intelligence (AI) techniques. The aim of this paper is to provide a comprehensive review of supply chain literature that addresses problems relevant to SCRM using approaches that fall within the AI spectrum. To that end, an investigation is conducted on the various definitions and classifications of supply chain risk and related notions such as uncertainty. Then, a mapping study is performed to categorise existing literature according to the AI methodology used, ranging from mathematical programming to Machine Learning and Big Data Analytics, and the specific SCRM task they address (identification, assessment or response). Finally, a comprehensive analysis of each category is provided to identify missing aspects and unexplored areas and propose directions for future research at the confluence of SCRM and AI.

Abstract:Sustainable supply chain management is a topical area which is continuing to grow and evolve. Within supply chains, downstream distribution from producers to customers plays a significant role in the environmental performance of production supply chains. With consumer consciousness growing in the area of sustainable food supply, food distribution needs to embrace and adapt to improve its environmental performance, while still remaining economically competitive. With a particular focus on the dairy industry, a robust solution approach is presented for the design of a capacitated distribution network for a two-layer supply chain involved in the distribution of milk in Ireland. In particular the green multiobjective optimisation model minimises CO2 emissions from transportation and total costs in the distribution chain. These distribution channels are analysed to ensure the non-dominated solutions are distributed along the Pareto fronts. A multi-attribute decision-making approach, TOPSIS, has been used to rank the realistic feasible transportation routes resulting from the trade-offs between total costs and CO 2 emissions. The refined realistic solution space allows the decision-makers to geographically locate the sustainable transportation routes. In addition to geographical mapping the decision maker is also presented with a number of alternative analysed scenarios which forcibly open closed distribution routes to build resiliency into the solution approach. In terms of model performance, three separate GA based optimisers have been evaluated and reported upon. In the case presented NSGA-II was found to outperform its counterparts of MOGA-II and HYBRID.

This article contributes to location-routing literature on three inter-linked aspects viz., formulation of a novel integrated low-carbon/green location-routing model for the demand side of a Supply Chain (SC) with a single product and multiple consumers, i.e., Drop-off Points (DoPs), a novel and robust solution approach through a Design of Experiment (DoE)-guided Multiple-Objective Particle Swarm Optimisation (MOPSO) optimiser and exhaustive analysis of the location-routing solutions (i.e., prioritisation, ranking and scenario analysis). The total costs, CO 2 emission and the traversed distances of the vehicles during transportation are optimised. The optimisation model for the strategic decision-making is formulated by effectively integrating the 0-1 mixed-integer programming with a green constraint based on Analytic Hierarchy Process (AHP). Due to the computationally NP-hard characteristic of the model a systematic and technically robust DoE-guided solution approach is designed using a commercial solver -modeFRONTIER ® .DoE guides the solution through the MOPSO optimiser in order to eliminate the un-realistic set of feasible and optimal solution sets. A popular multi-attribute decision-making approach, TOPSIS, evaluates the solutions found from the Pareto optimal solution space of the solver. Finally decision-makers' preferences are analysed for monitoring the changes in the controlling parameters with respect to the changes in the decisions. A scenario analysis of the location-routing events by considering alternative possible outcomes is also conducted. It is found that the implemented methodology successfully routes the vehicles with optimal costs and low-carbon emission thus contributing to greening the environment on the demand side of a SC network.

This chapter considers the importance of decision support systems for supply chain risk management (SCRM). The first part provides an overview of the different operations research techniques and methodologies for decision making for managing risks, focusing on multiple-criteria decision analysis methods and mathematical programming. The second part is devoted to artificial intelligence (AI) techniques which have been applied in the SCRM domain to analyse data and make decisions regarding possible risks. These include Petri nets, multi-agent systems, automated reasoning and machine learning. The chapter concludes with a discussion of potential ways in which future decision support systems for SCRM can benefit from recent advances in AI research.

This article presents an effective solution method for a two-layer, NP-hard sustainable supply chain distribution model. A DoE-guided MOGA-II optimiser based solution method is proposed for locating a set of non-dominated solutions distributed along the Pareto frontier. The solution method allows decision-makers to prioritise the realistic solutions, while focusing on alternate transportation scenarios. The solution method has been implemented for the case of an Irish dairy processing industry's two-layer supply chain network. The DoE generates 6,100 real feasible solutions after 100 generations of the MOGA-II optimiser which are then refined using statistical experimentation. As the decisionmaker is presented with a choice of several distribution routes on the demand side of the twolayer network, TOPSIS is applied to rank the set of non-dominated solutions thus facilitating the selection of the best sustainable distribution route. The solution method characterises the Pareto solutions from disparate scenarios through numerical and statistical experimentations. A set of realistic routes from plants to consumers is derived and mapped which minimises total CO 2 emissions and costs where it can be seen that the solution method outperforms existing solution methods.