Showing papers by "Chandrasekharan Rajendran published in 2021"

Corporate social performances of firms in select developed economies: A comparative study

Abstract: Corporate Social Performances (CSP) has a determining role in the Environmental, Social, and Governance (ESG) scores of firms. Corporate Social Responsibility (CSR) strategy scores in the ESG ratings can provide a measure of the CSP concerts of firms. We observe in this research, the mean differences in the CSR strategy scores and the ESG scores of firms in select developed economies such as; US, UK, Japan, and Australia, representing different geographical regions globally. Thomson Reuters ESG scores based on ten major parameters and over 400 company level indicators are used to empirically evidence the study. The initial data of average performances on ESG indicators of 939 firms considered for a period of five years from 2014 to 2018 is analyzed. The results imply that the mean differences in the CSR strategy scores are not significant, considering the developed economies, deliberated in the study. Along with that, we observe a significant mean difference in the ESG scores of Australian firms considered for the study, in comparison with the firms from other developed economies. And the study confirm that the CSR strategy scores are significant predictors of ESG performance scores of firms in the developed economies considered for study.

Supply Chain Risk Management in Indian Manufacturing Industries: An Empirical Study and a Fuzzy Approach

Abstract: Risks are inevitable in supply chains, and they need to be detected early and appropriately addressed. This chapter primarily attempts to identify early warning signals and implement suitable mitigation decisions to meet exigencies related to the management of supply chain risks. Further, the chapter also presents an approach that addresses the risks in an integrated manner. First, a framework is essential to understand the relationships between the independent and dependent variables. An empirical study is undertaken by developing a questionnaire that captures the perceptions of the supply chain practitioners on risks perceived in their supply chains, and the framework is subjected to validity tests. Secondly, the data obtained from these surveys is utilized to develop a fuzzy model for identifying and predicting all plausible risks based on the instantaneous risk vector. Fuzzy Cognitive Map (FCM) is used to represent the overall behavior of the dynamical system of the supply chain. The instantaneous risk vector is passed on to the dynamical system to identify all plausible risks that may appear in near future. The resultant vector obtained suggests that ignoring the initially perceived risks eventually lead to possible disruptions in the supply chain. The resultant vector thus obtained is useful for decision making to alleviate the impact of various types of risks. Finally, the relative comparison of the mitigation strategies’ ranking was made for the results obtained from regression, FCM, and Fuzzy TOPSIS.

tofee-tree: automatic feature engineering framework for modeling trend-cycle in time series forecasting

Abstract: Most time series forecasting tasks using Artificial Neural Networks (ANNs) relegate trend-cycle modeling to a simple preprocessing step. In this work, we propose an automatic feature engineering framework for modeling the trend-cycle (tofee-tree) in time series forecasting. The first stage of the framework automatically creates over 286 deterministic linear and nonlinear engineered features to model the trend-cycle. These features are based only on the time of observation and length of the time series, making them domain-agnostic. In the second stage of the framework, a SHapley Additive exPlanations (SHAP)—based feature selection procedure using Light Gradient Boosted Machine (LightGBM) selects the most relevant features. These relevant features can be used for forecasting with ANNs in addition to the auto-regressive lags. Two popular ANNs—Multi-Layer Perceptron (MLP) and Long Short Term Memory network (LSTM) are used to evaluate our proposed tofee-tree framework. Comparisons against two empirical studies using the M3 competition dataset show that the proposed framework improved the overall Symmetric Mean Absolute Percentage Error (SMAPE) in the one-step, medium- and long-term. The relative improvement in one-step SMAPE is 3% for MLP and 23% for LSTM. We also show that the residual seasonality left after deseasonalization can be modeled using the tofee-tree framework.

A Comparative Study on Classical and New Hybrid Continuous-Review Inventory Ordering Policies in a Supply Chain Using Mathematical Models

Abstract: One of the most challenging tasks in managing a supply chain is to select an efficient inventory ordering policy, i.e., deciding on ‘when to order’ and ‘how much to order’ while minimizing the total cost and maximizing the service level. Classical inventory policies based on periodic and continuous-review are generally implemented in practice. In our work, we attempt to combine the characteristics of both periodic-review order-up-to (R, S) policy and continuous-review (s, Q)/(s, S) policies to propose two new hybrid ordering policies, namely continuous-review (s, Q*) and continuous-review (\({s,\mathrm{O}\mathrm{Q}}^{*})\) hybrid policies. We further develop mixed integer linear programming (MILP) models to obtain the optimal policy parameters by considering a single-stage and two-stage supply chain with discrete, deterministic demand over a finite planning horizon. The proposed policies are benchmarked against the existing order policies, namely (R, S), (s, Q), (s, S) and hybrid (R, S, Qmin) policies. From results, we observe that the performance of the proposed hybrid policies outperforms the existing classical and hybrid policies in terms of total supply chain cost.

Bounding strategies for obtaining a lower bound for N-job and M-machine flowshop scheduling problem with objective of minimising the total flowtime of jobs

Abstract: In this paper, bounding strategies for determining a lower bound on the completion time of a job sequenced in each position in the permutation sequence on each machine in permutation flowshop scheduling problem with minimisation of total flowtime of jobs as objective are discussed Basically, the bounding strategies are machine-based bounding strategies used for determining the lower bound on total flowtime of jobs for all the small-sized and large-sized benchmark flowshop scheduling problem instances proposed by Vallada et al (2015) The lower bound matrix can be pruned as tightening constraints into the mixed integer linear programming (MILP) model with objective of minimisation of total flowtime of jobs Since the flowshop scheduling problem with total flowtime objective is difficult, two kinds of linear programming (LP) relaxation methods are used for determining an LP-based lower bound on total flowtime of jobs for some benchmark problem instances proposed by Vallada et al (2015)

Mathematical Models and Heuristics for an Inventory Routing Problem Without Split Deliveries

Abstract: We consider a finite horizon Inventory Routing Problem in which the vendor monitors the inventory at the retailers and makes replenishment decisions for them. A single vendor manages several retailers who face deterministic but dynamic demands. The vendor decides the delivery schedule, delivery quantities, and vehicle routes that minimize the total supply chain cost. For practical purposes, we assume that split deliveries to a retailer are not allowed and that the backorders are allowed at the retailers. Two mixed integer linear programming models are proposed to solve the specified Inventory Routing Problem (IRP). Since IRP is NP-hard, two heuristics are also developed for the same problem. Numerical experiments were conducted on randomly generated data instances to evaluate the proposed models. The numerical study reveals that the proposed models are computationally efficient. We observed that the optimality gap of the solutions from the mathematical models was no more than 3% in all the tested cases. We also show that the proposed heuristics provide consistent near-optimal solutions. The proposed models and heuristics could serve as efficient decision-making tools to help supply chain managers make quick tactical level decisions.

Optimal and Heuristic Profit Sharing Using Sales Rebate Contract in a Multi-level Supply Chain

Abstract: A supply chain (SC) is an interdependent network of many SC members who add value to the product successively until it gets to the final stage and attains utility. One of the most important problems in any supply chain (SC) consisting of organizations is the sub-optimization due to the fact that the SC decision making is distributed over various players. An interesting solution to the issue, which leaves the decision power and supply chain structure intact, is the use of contract mechanisms. Ideally, contract mechanisms ensure that the whole SC is optimized as if it were a single unit (coordination) and is designed such that all the players benefit from working together through the coordination mechanism (win-win). Once coordination is in place, the next challenge is to devise some mechanism to share the profits gained due to coordination. To enable coordination among the SC members, sales rebate contracts in a multi-level SC using fair share mechanism have been devised. In the sales rebate contract, the supplier charges the buyer wholesale price per unit purchased but then gives rebate to the buyer per unit sold above some threshold. The objective of this paper is to determine the optimal rebates given to downstream SC members and threshold value of sales, which maximizes the SC profit as well as fairly shares the profits among the SC members. This objective can be achieved by minimizing the deviation between proportion of each member’s value addition and proportion of each member’s respective share in the total profit by coordinating through sales rebate contract. An analytical model is developed to formulate the supply chain coordination problem and simultaneously formulate the optimization problem to achieve fair distribution of profit among the SC members. Three heuristic search techniques based on genetic algorithm and simulated annealing have been proposed to solve the problem.

CARIMO - A heuristic approach to machine-part cell formation

Abstract: This paper presents a correlation analysis-based heuristic for the machine-part cell formation in the context of cellular manufacturing systems. Two new indices, viz. “mean correlation index” for forming the part families and “relevance index-modified” for identifying the appropriate machine cells are proposed. The machine-part cells formed by the proposed heuristic resulted in a higher grouping efficacy (GE) for 14.3% of the test instances gathered from the literature, and it performed equal to the best in class heuristics available in the literature for 80% of the test instances. The method presented in this paper has set a new benchmark GE for 5 of the 35 test instances used by the researchers in the context of machine-part cell formation without singletons.

An Integrated Problem of Production Scheduling and Transportation in a Two-Stage Supply Chain with Carbon Emission Consideration

Abstract: This study considers a scheduling model for a supply chain system that can choose a set of subcontractors from the available subcontractors (non-identical manufacturing facilities) to fulfill a part of its orders/jobs in order to maximize the supply chain profitability. Further, this study aims to reduce carbon emissions from transportation activities in the supply chain. The orders/jobs to be processed have different processing times on different manufacturing facilities and due dates. All the completed jobs at the outsourced centers (sub-contractors) need to be transported back to the central manufacturing facility. The present study integrates three issues: (1) selection of subcontractors; (2) scheduling of jobs, and (3) logistic scheduling with carbon emission consideration; all these decisions in a supply chain have not been considered together in the existing literature. We present this problem with the objective of minimizing the total costs associated with production and logistic decisions, and propose a mixed-integer linear programming model.