Applications of Emerging Technologies and AI/ML Algorithms

Due to globalization of markets, products are moving throughout the country and exported. Quality monitoring and traceability in a supply chain is hence essential. Blockchain technology ensures the privacy or resistance to data immutability of a decentralized database in case an attacker tries to alter or delete data inside the blockchain. The purpose of our research is to develop a holistic approach to make a perishable food supply chain resilient by ensuring a secure way of storing and monitoring supply chain data and identifying and optimizing the relevant factors that lead to bacterial contamination in food supply chain under uncertainty. We developed a blockchain interface to add, validate, and send data related to detailed wheat supply chain from farmers growing crops to harvesting, storing in the silos or elevators, then processors for milling and sifting to develop end products (bread, biscuit, buns, tortillas, etc.) and distribute to retailers who finally sell them to consumers. After the data is securely stored in the decentralized ledger, in our next step, we came up with a data-driven technology such as machine learning (Q-learning model) to filter relevant parameters impacting food quality at all stages of the supply chain and optimize them to ensure good quality.

The present work has been carried out in the logistics distribution of an automobile manufacturer in South Africa for improving their outbound vehicle logistics distribution efficiency. As far as we know, the literature on the distribution and logistics systems used in the automotive industry does not currently contain a Mixed Integer Linear Programming (MILP) model for implementing First In First Out (FIFO). This study presents the development of a comprehensive shipment plan for the automobile manufacturer to (i) deliver finished cars from the manufacturing plant to other facilities and (ii) adhere to a range of loading, routing, and capacity constraints while meeting the vessel-loading timetable and shipment plan with the minimum transportation and inventory costs. The model accounts for dynamic parameters such as train and trailer availability, production volume, ship timetable, and demand at the ports of South Africa. An MILP model is developed to capture the business constraints and the time-dependent storage costs at the facilities, which vary depending on the number of days cars are stored in the facilities. The proposed MILP model uses FIFO strategy to ensure that finished cars leave port warehouses in the same order that they arrived. The model offers a cost-effective shipment plan to meet the vessel demands and the daily transportation mode requirements. These results will benefit the decision-maker in automating the shipment plan.

Public procurement is described as a dynamic, real-time negotiation between a purchasing association and several pre-qualified suppliers contesting against one another to achieve the opportunity to deliver goods or services to the purchasing organization. A significant number of enterprises use Reverse auction (RA) for supplying products. This paper investigates aspects of the RA system using the optimization model to keep up the work on comprehending the characteristics of the impact of performing RA on the bidding results for multiple products. In particular, a graph-based model has been developed for the RA bidding framework to evaluate the consequences of the bids collected in public procurement. Our method aims to confirm the determination, whether procurement is profitable based on the previous auction datasets. The analysis involves a scenario consisting of companies that will dominate the market. This study also consists in determining the decision to occur an auction.

Digital transformation is changing the business landscape; hence Digital technology integrates suppliers, customers, regulatory agencies, stakeholders and other related parties to provide solutions to the problems faced by supply chain information. This research, therefore, aims to analyse the impact of e-Procurement and blockchain on government public bidding in China and further explores the obstacles faced in implementing and integrating technology into their systems. Through interviews with government employees and suppliers, this research analyses the application performance and challenges of e-Procurement and blockchain in procurement and the benefits to downstream supply chain finance. The results show that e-Procurement and blockchain are handy tools in improving the process of government public bidding and enhancing the transparency of the procedures and the speed of information flow. They help downstream suppliers to get more financial support and help the government procurement industry to develop better. A transparent market among customers, retailers, suppliers and manufacturers in the supply chain allows companies and governments to achieve a win–win situation.

This study aims at designing a distribution network for public procurement of fresh fruits and vegetables in India. The present model is conceptualized in an omni-channel environment to offer a seamless buying experience to consumers through government-regulated markets. It addresses the key challenges of fresh food loss due to quality degradation and freshness-keeping effort along with transportation network design during fresh food distribution. Hence, a cost optimization model is developed which minimizes the cost associated with transportation, quality loss, and freshness-keeping effort using a mixed-integer nonlinear programming model. The proposed model is solved using an exact solution approach in a pyomo environment with the CPLEX solver for different problem sizes. The effects of quality loss cost and freshness-keeping effort on the total cost and freshness-keeping effort decisions were also analyzed.

With the objective of minimizing the total distance travelled, the work proposes a novel procurement-distribution design network of the supply chain for the fair price shops (FPS) in the public distribution system of India. The work is a two-step process. The first step is to understand the existing distribution system. This starts from identifying the existing location of the FPS and converting them to its geographical locations in terms of longitudes and latitudes which are converted to distances to estimate the total distance travelled. The study, as a case, is conducted for one of the districts of Kerala, which is the rice bowl of Kerala which has 14 depots that serves the total 941 FPS in the district. The total distance travelled by the current distribution network is identified in the first step. The second step of the work is to determine the optimal number of depots that would be required to serve the existing 941 FPS of the district. To determine the optimum number of depots K-mean clustering applying the centroid approach is proposed to be adopted. We are proposing a novel distribution cum procurement supply chain. The trucks on delivery of the goods can collect the grains procured at the FPS from the farmers. The second step would primarily optimize the number of depots required to serve the FPS, and then identify the FPS that could be distributed from the given depot. The objective is to provide a scientific solution to minimize the total transportation distance by identifying the geographical location and adopting the k-mean clustering procedure. Finally, the benefits of implementing the proposed model will be quantified and discussed in detail by comparing the existing distribution system.

As we move further into the twenty-first century, technology has played an essential role in people’s lives inevitably, and generated massive information called ‘Big Data (BD)’. With the ability to manage massive dataset, big data analytics can usefully extract the insight from big data and support the firm to leverage decision-making. Hence, the interest in big data applications has spread to comprehend many areas of study, including Supply Chain Management (SCM). However, the academic articles studying the employment of Big Data Analytics (BDA) in SCM are limited. Besides, most of those academic papers offer less interest in the entire SC systems. Most of them prefer to study in an individual SC area. Thus, this study aims to investigate state of the art in this domain through Systematic Literature Review (SLR) and discuss future research opportunities. We found that optimisation, simulation and visualisation tend to be the most appropriate BD tools to apply in SCM. Also, linear programming, statistics, association rule mining, fuzzy logic and decision tree are likely to be the most suitable BDA techniques for SC operations.

This paper proposes and demonstrates (a) a supervised machine learning methodology to predict the expenditure incurred on procurement, repair, and reconditioning of components as well as expenditure incurred on procurement of fuel based on the performance of the Indian Railways and (b) an unsupervised machine learning methodology to classify the good and poor performing administrative zones, using the data of expenditure incurred on procurement, repair, and reconditioning of components as well as expenditure incurred on procurement of fuel and the performance. The first methodology will aid managers in determining whether the expenditure incurred is more than what should be incurred. Further, it may also benefit the managers in fine-tuning the frequency of replacement of components. The second methodology will assist managers in searching for best practices of maintenance in good-performing zones, which can be propagated in the poor-performing zones to lower the overall expenditure on maintenance.

In this paper, we present a review of Blockchain based technologies and their application in supply chain procurement. The blockchain has two main characteristics that may be leveraged for adoption in supply chain networks and logistics. First, the exchange of information in blockchain is real time, secure, verifiable, trustable and these are accessible to all members of the network. 3D Printing is an excellent application for Blockchain based procurement. In Blockchain based 3D printing, the aim is to secure contracts from suppliers that may be used in assemblies or for spare parts. An OEM can issue a call for proposals in real time from 3D printing suppliers. The bids can be received securely using the blockchain with partial/complete order quantities. In this paper, we show how the blockchain can be used to integrate the four stages of 3D printing procurement: Scan and Design, Build and Monitor, Test and Validate, and Deliver and Manage. Finally, we present a case study for scheduling 3D printers in manufacturing network. We examine a 3D printer scheduling problem by considering multiple printing alternatives in which various types of products can be printed in different forms using Mixed-Integer linear programming (MILP) under scenario-based experiments.

A typical problem faced in any supply chain is the simultaneous planning of inventory and logistics. In this work, we consider a representative problem of clustering, lot size optimization, and transport planning faced in many logistics industries. It is a problem of an inbound logistics system for which demand is generated through the outbound logistics. The demand for parts for a finite time period is given and is deterministic. The demand occurring at a warehouse is met by procuring raw material from the suppliers that are spread out geographically. The order pickup is done using heterogeneous trucks which start from the warehouse, pick up the order from the supplier, and return to the warehouse. The entire trip should be completed within the given lead time without any permissible backorder or delay. The problem can be described as Constrained In-Bound Multi-Period Multi-Part Inventory Routing Problem with Heterogeneous Vehicles (CIBIRPHV). The problem is distinct and different from the Inventory Routing Problem present in the literature. The objective is to minimize the total costs, comprising of transportation costs, inventory carrying costs, and warehouse storage costs. We propose a mathematical model to first cluster the suppliers, followed by another mathematical model for order planning and pickup.

Roads transport lions-share of the freights as they can reach each and every corner of the country. In road freight transport, truck is more preferable due to its advantages over railways due to its ability to reach beyond geographical barriers, quantity barriers, and ease of availability of trucks. In order to achieve efficient and reliable operations of the trucking industry the truck drivers play a vital role. As one of the major issues the trucking industry facing is shortage of truck drivers. The objective of this study is to identify the performance measures and efficiency calculation. In order to this study adopted and hybrid approach by combining Balance Score Card (BSC) and Data Envelopment Analysis (DEA). The present study considers a case of Indian trucking industry, and the efficiency of truck drivers has been calculated. This study helps the organization in reciprocal learning, by focusing on the specific factors or perspectives for efficiency improvement. The results of the study can also be used in strategic implementation to improve the performance of the trucking industry.

The movement of commodities, services and information in the pharmaceutical industry should be planned to effectively turn raw materials into completed dosage forms of medications. Pharmaceutical manufacturing companies frequently purchase huge amounts of active pharmaceutical ingredients which require efficient procurement systems to maintain everyday supply and activities. One of the major goals of pharmaceutical supply chain management is to efficiently apply information technology to its procurement systems. This research examines the Indian pharmaceutical companies’ managers’ perspectives about e-procurement technologies through a qualitative study. Based on extensive literature review and the obtained results, a research model for examining the pharmaceutical companies’ behaviour to use e-procurement technologies is proposed. The proposed model is an extension of the Technology acceptance model. The identified factors for the model are data integration, performance expectancy, cost and user benefit, innovation, infrastructure, operations, individual attributes, perceived usefulness and perceived ease that will influence pharma companies’ attitude-intention-behaviour to use e-procurement technologies.

To evaluate the economic risk, a novel supply chain network is designed, which is a multi-period, multi-echelon in nature. The dynamic profit is evaluated in terms of net present worth (NPW), involving fluctuating money’s value with time and depreciation to manage demand–supply ratios. The designed mixed integer nonlinear programming (MINLP) supply chain model incorporates technical aspects of potential site locations for manufacturing, import, inventory, and retailer, along with feasible and flexible connectivity for distribution. Economically, detailed capital and operating cost expenses are evaluated to find the cost distribution. Environmentally, greenhouse gas emissions are calculated throughout the different phases of the model’s life cycle via life cycle assessment (LCA) following the government’s carbon emission injunctions for the penalty of $35 per ton carbon equivalent. The SC feasible and infeasible operating conditions are identified via sensitive parameters to assist the overall strategic decision. Here, the sensitive parameter, specifically the transport cost below ₹ 3.25 per km per kg, is found to be a crucial parameter. When unit transport cost increases by 116.67%, the entire project becomes infeasible and risky, making NPW negative. Overall, the model incorporates the technical, economical, and environmental aspects as a step toward a sustainable supply chain by taking the cash crop cotton as the case study product for Pune city.

Most of the manufacturing firms are exposed to the risk of Commodity price volatility which can have a substantial impact on their operational and financial decisions. One of the most used metal commodities in manufacturing industries is Steel and the known fact is that steel prices vary over the time making it difficult for procurement decisions. With the current advancements in the field of Artificial intelligence and Machine Learning, there is a growing emphasis on development of accurate forecasting methods for commodity prices that plays an important role in procurement decisions for manufacturing firms. The current paper aims to develop ML-based forecasting models by employing Tree-based algorithms namely Regression trees and Random Forests to forecast steel prices. Past 10 years monthly historical data of several variables impacting the prices of Steel and the prices of steel are considered for forecasting Steel Prices. The results reveal that the proposed methods present a promising forecast with high accuracy. The major performance metric used in this study to measure forecasting accuracy is Mean Absolute Percentage Error (MAPE). The tree-based models used in this paper gave MAPE of less than 5%, indicating that they outperform other traditional forecasting approaches used in the literature, making purchase decisions easier.

During COVID-19, many large economies like European Union, the U.K., Japan and Mexico officially encouraged companies to undergo coopetition to counter commotions, hoardings and stock-piling. Observing the benefits, many companies and economies are formally coming up with official norms for coopetition. This study suggests a novel interactive strategic algorithm using Game Theory to model the pricing decisions and quantify the optimal order quantity and profit during coopetition. For the analysis, we assumed a dual-channel supply chain consisting of the manufacturer, retailer and e-tailer. The Stackelberg game was used to model the interaction between the upstream and downstream partners. We assumed coopetition between the downstream retailer and e-tailer and modelled the interaction. Later we analysed the performance of the channel partners under normal buying conditions and panic buying and compared them to derive the propositions. During the coopetition model, the retailer’s optimal price, order quantity, and profit increased during panic buying, whereas only the optimal price increased for the e-tailer. Manufacturer could reap maximum benefits during the panic buying period. The interactive analytics developed can aid the management practitioners in developing decision support systems and multi-agent systems during coopetition.

The most difficult predictive challenge in supply chain disruption management is order delivery delay. Identifying the risk in delivering an order in the scheduled time will help the company to focus on the prioritized orders to mitigate the disruption before its occurrence. This research presents a machine learning-based predictive model for delivery risk prediction of different product orders. The proposed approach deals with an imbalanced class problem, where the frequency of orders which have the delivery risk is rare when compared to the orders that do not. The Area Under the Curve (AUC) score is the selected performance metric for the proposed risk prediction problem. With a comparative analysis, it is found that the Random Forest model in Synthetic Minority Over-sampling Technique (SMOTE) with the Tomek link gives a better performance with an AUC score of 0.80. It is also found that the Random Forest model performs better in SMOTE and SMOTE Tomek oversampling methods, whereas K-Nearest Neighbour (KNN) performs well in the random oversampling technique.

This paper has analyzed the effects of massive public procurement of food grains and their indirect impact on the environmental front. For decades, the practice of food grains procurement has created a barrier to a more equitable and sustainable procurement. We have applied the concept of system dynamics to analyze the critical loops that played a significant role in the current scenario. We have proposed equity as an alternative approach to counter the skewed procurement and sustainability issue.

Real-time controlling the vaccine shortage and wastage within the supply chain network (SCN) is a critical challenge for healthcare professionals. However, in the literature, only limited studies are available on theoretical support and application of 5G to control the shortage and wastage of vaccines and no effective framework has been developed. Moreover, we are proposing a framework for a 5G enabled vaccine supply chain (VSC) digital twin to control the shortage and wastage of vaccines across the SCN. The 5G technologies enhance real-time visibility and connectivity of the supply chain (SC) at a highly granular level. This paper aims to propose an application of 5G technologies to provide the digital twin framework for improving visibility and connectivity that will help to make the real-time decision for controlling vaccine shortage and wastage within the SCN. As a result, this proposed digital twin framework for the vaccine supply chain will help improve delivery performance during pandemic and emergency times through highly granular visibility and connectivity across the network.

In this paper, we explore the issues in supply chain procurement and distribution of plantation crops. Specifically, we explore challenges related to visibility and information sharing related to production data, problems in price disparity, supply–demand mismatch, and procurement auctioning. The context is analyzed under the lens of a case study on rubber crop plantation in the state of Kerala. We use simple comparison and review to explore the scope of applicability of data analytics and AI-ML for superior performance outcomes. The paper concludes with a discussion on implications to practitioners, limitations, and future scope of research.

Products including garments, clothing accessories, and consumer electronics are prone to suffer from short product life cycles as a result of fast technological advancements, manufacturers' marketing methods, and rapidly changing customer tastes. With the growing demand for the technology-driven product, the mobile phone's lifespan has fallen to three to six months. To compete with the global market, the mobile manufacturer cut down the price of mobile phones and provides upgraded specifications. For this reason, consumer wants to experience the new technology in every instance and exchange their mobile phone for a newer one. The performance of the mobile phone does not reduce, but to upgrade to modern technology, the customer intends to swap it with the fresh one. For this purpose, a circular economy comes into the picture. Remanufacturing is becoming a critical element of a circular economy where products are created, produced, used, and retrieved to avoid any kind of waste and decrease the extraction of raw materials. However, lots of studies are carried out on the quality grading of the returned core, but we did not find any of the quantified definite indexes for the quality of the yielding core. Thus, in these articles, we attempt to quantify the quality of returned core (mobile phone) by using MCDM and PCA / FA methods. Finally, the reduction equation is established to predict the quality of returned core, which reduces the remanufacturing cost by providing optimal shorting.

Small firms, also known as SMEs or MSMEs, face acute shortage of capital. These firms have huge pending receivables from large corporate buyers and government agencies which contributes significantly to this shortage. Despite a series of initiatives by the government, the issue remains critical to the survival and growth of MSMEs in India. This study explores the current issues and gaps in MSME receivables realization and government interventions in India, and proposes an integrated Blockchain-based architecture for end-to-end receivables management of Indian MSMEs using Ethereum. This will enable timely, cost-effective, transparent and competitive receivables realization for the small firms freeing up a significant amount of capital.

Demand variations, disruptions, and environmental regulations stand as major setbacks in the perishable food supply chain (PFSC). Hence, in PFSC there is a need for identification of enablers and critical enablers that can help for better decisions to overcome the above mentioned issues by making the supply networks more resilient. In this chapter, first with the identified enablers the most influential ones have been recognized with the multi-criteria decision-making (MCDM) based Decision making trial and evaluation laboratory (DEMATEL) method. In addition to the ranking of enablers, we have determined the cause and effect relationship of the enablers for the perishable food supply chain using DEMATEL. Later, a hybrid MCDM i.e., Analytical Hierarchy Process (AHP) - Preference ranking organization method for enrichment evaluation (PROMETHEE II) has been used to rank the enablers and further identify the key enablers for the considered PFSC. The results obtained from the above models provide the critical enablers which further help the members in the supply chain be more resilient.

Maintaining transparency in the perishable goods supply chain is critical to ensure that the items are not tampered with and that customers have faith in the products. This work aims to solve this problem utilizing blockchain technology, which is well-known for its decentralization, immutability, and trustless nature. The research goal was to develop an Android application that is both cost-free, easily deployable, and easy to use for recording product data and the supply chain. Before purchasing a product, a consumer might use this app to scan the QR (Quick Response) code to retrieve all of the product’s details and supply chain to ensure that the product has not been tampered with and is still within its shelf life.

Classification is a supervised machine learning technique that is used to predict the class or category of a new observation based on training data. Classification techniques can be broadly categorized into logic-based (decision trees and rule-based classifier, perceptron-based (neural networks), and statistical techniques. In this study, we focus on developing a mathematical model based heuristic for binary classification problems. The application of Operation Research (OR) techniques is quite rare in the literature on classification. We propose a novel Linear Programming (LP) based model for binary classification. A very popular dataset available in the literature, Iris, has been used in our analysis. It is a multi-variate dataset first used by Fisher in [1], which consists of three classes of fifty instances each, where each class refers to a type of Iris plant. The classes are as follows: Iris Setosa, Iris Versicolor, and Iris Virginica. The proposed Linear Programming model consists of additional variables to show the interaction effect between the predictor variables along with the actual predictor variables so that non-linearly separable data points can also be classified by using the proposed LP based binary classifier. The values of the decision variables, obtained from building the classification model using LP on training data set, are used to predict the class of observations on the test data set. The overall accuracy of the proposed 2-stage classifier is found to be comparable with the results reported in the literature.

One of the most challenging tasks is forecasting intermittent demand, yet the most important activity in the forging industry since it serves as a foundation for production and inventory level planning. It is likely to be the most difficult task in handtools manufacturing as well. While working with these kinds of demands, exponential smoothing is frequently utilised in practice. More improved approaches, such as the Croston method, SBA, MA, ARIMA, SARIMA model, and so on, have been researched based on the exponential smoothing method. Demand unpredictability and intermittency offer obstacles in accurate demand forecasting using traditional or better methods. Support vector machine (SVM) models have been found to outperform previous models in terms of accuracy. However, there are certain drawbacks to basic SVM models, such as the fact that the computation takes more time and does not result in a statistically significant gain in accuracy, and there are a few reasons for model resilience in demand forecasting. To anticipate intermittent demand, we used an adaptive univariate SVM (AUSVM) model. Real-world data from the forging (handtool) sector indicates its performance when compared to current models. In terms of computational time, AUSVM clearly surpasses basic SVM. The computational findings of the forging industry handtool scenario show that for a lot of non-smooth demand series, AUSVM provides an analytically important increase in accuracy and best inventory performance when compared to very famous parametric models. The paper concludes with an explanation of why AUSVM is better for forecasting demand and inventory control in the forging industry, in general, and for a handtool manufacturer in particular.

This research study will conduct a literature review on the development of electric vehicles and charging station infrastructure. We used the four-phase method to review the literature. The review process includes bibliometric search, descriptive analysis, scientometric analysis, and citation network analysis. In phase I, the 957 articles retrieved from Scopus and Web of Science from 2008 to 2022 were systematically screened. The final selected articles were then subjected to descriptive analysis to identify the most influential authors, articles, keywords, and countries in the EV research domain. The research concepts/themes and methods were then classified using thematic analysis. Numerous discoveries have been made in the development of electric vehicles and charging infrastructure as a result of this review. China, the United States, and Germany are the leading countries in all areas of EV and EVC research. The research gap and issues of EV and EVCS are highlighted at the end of the review, as is the scope for future discussion.

The contract management process is a tedious and often manually performed task in the procurement process in the industry. It includes requesting bids, negotiating, drafting the contract, etc. Within this purview, there is ample opportunity for smart automation. In this paper, we present a general framework for a negotiating bot that can bargain with the vendor for an agreement on the price in the contract negotiation process. Based on the data provided by the vendor, the bot determines a fair price as per market standards, using which the user (procurement staff) can further negotiate for a better deal as per the systematic steps described by the decision bot. This method can benefit the companies by saving a significant amount of time and money for the organization and reducing human dependency on such routine tasks.

The outcome of construction activities leads to the production of large amounts of solid waste, primarily known as construction and demolition (C&D) waste. The reutilizing of C&D wastes plays a vital role in the sustainable growth of the environment, economy, and public health. The existing recycling methods have limitations, such as cost, human intervention, unstable identification process for recycling, on-site sorting techniques, irregular landfill events, and a lack of an effective waste tracking system. The paper proposes an end-to-end improved convolutional neural network (EEI-CNN) based reverse logistics model for recycling C&D waste to overcome these issues. The EEI-CNN is a customized convolutional neural network that performs the classification of C&D waste aggregates. The refine the efficacy of EEI-CNN, a preprocessed image is used. The effectiveness of the proposed method is judged for an openly available C&D waste image dataset. The evaluation metrics like accuracy, precision, true positive rate, true negative rate, and F-score are estimated. The proposed method outperforms existing methods based on comparative analysis.

Today's global business is evolving into a one-economy and one culture. As a result, the production sector is shifting toward high-value-added operations, and supplier and risk management needs may change significantly. The current overview addresses the issues of finding suitable and viable suppliers, prioritization, and risk management in procurement. The main types of supplier selection have been determined at this point. Companies utilize qualifying criteria, selection criteria, and other considerations to categorize criteria during the selection process and risk management. The SLR technique is utilized to establish the relative relevance of supplier prioritization and risk management in procurement. In the context of the Indian manufacturing business, this helps to determine purchaser preferences in prioritizing suppliers. A review was conducted for this aim to define supplier prioritization and proactive risk management in procurement. This review paper's contribution is to examine the research findings and future research directions. The SLR study helps find and contribute to viable suppliers based on delivery, quality, cost, added value, flexibility, risk, service, ecological ways, and social accountability, as well as operational and disruption risks in risk management in procurement. Researchers can use an innovative approach to operations research to solve supplier prioritization and risk issues in procurement uncertainty concerns.

Most businesses nowadays are designing their products and services with the customer in mind. Many businesses throughout the world are expected to shift from a product-centric to a customer-centric attitude. Customer relationships, experiences, and happiness are therefore crucial for any business's long-term existence, sustainability, and profitability in any industry, yet small businesses lack the resources and skills to succeed. An integrated decision-making framework is built in this present study, integrating diverse data mining methodologies from many fields. The primary goal of this research is to improve the application of predictive analytics in small and medium-sized organizations. The decision-making framework in the form of a Customer Relationship Management (CRM) tool for an online retail sector is the solution offered as part of this study. An integrated decision-making framework tool is built in the pretense of a predictive analytical CRM system, with seven core characteristics and more than thirty sub-activities. The seven features created as part of this study are Data Visualization and Analysis, Customer Segmentation, Customer Classification, Product Recommendation, Customer Linked Predictions, Sales Forecasting, and Forensic Analysis, since they are frequently requested by CRM tool users. It is built using a variety of data mining and machine learning methods. The tool is then made available as a real-time online application. This tool, which consists of a frontend and a backend application, is essentially designed to give users a complete picture of the data. Aside from that, there are several enhancements that can be made to this tool.

Predicting top-performing companies is an essential part of long-term business-to-business relationships. For example, supplier selection, sourcing risk management, enterprise loan customer selection, and stock selection. Moreover, long-term financial decisions involve analyzing the microeconomies of the companies. In other words, ratio analysis helps us to diagnose the health profiles of companies to select loan customers, equities, suppliers, vendors, producers, buyers, sellers, and others, for any business collaboration. However, the financial data by its nature is noisy, and the relationships among fundamental variables are dynamic due to the external influence of the changing macroeconomies. This study reviews optimal sampling techniques used in machine learning models for mitigating the problems caused by changing economic environments. It brings some immensely important sampling designs to develop models for ration analysis in decision support systems. Filtering noise due to external disturbances necessarily imposes constraints on the industry, size, consistency, history, performance, and the environment of the company. The moving window system allows us to retrain the models with more recent environments. It makes the model comparison difficult, an essential step of model validation. However, in real-world situations, models must prove to be more profitable rather than more accurate for usability.

In order to get the market to give the public sector products and services that have the least possible negative impact on the environment, green public procurement, or GPP, is now being implemented. Despite the fact that the scope varies from nation to nation, it is a widely acknowledged environmental policy tool. In the meanwhile, modifications in environmental legislation have led to states randomly purchasing goods, services, and works as well as their application. Green Public Procurement (GPP) has been shown to be an effective strategy for achieving the environmental policy objectives indicated in the communications from the European Commission. The field of artificial intelligence (AI) has expanded dramatically at the same time that businesses, governments, and society have greatly benefited from the intelligence of computers with machine learning capabilities. They also have an impact on broader trends in global sustainability. Key problems for sustainable manufacturing can be helped by artificial intelligence (e.g., optimization of energy resources, logistics, supply chain management, waste management, etc.). In this framework of smart production, there is a push to integrate AI into green manufacturing processes in order to abide by increasing environmental requirements. The government mandates a seamless integration of a complex system of laws, institutions, innovations, health care, nutrition, and education into all goods and services. The bulk of ecological and biodiversity study domains, as well as environmental and ecosystem management in general, are anticipated to benefit from advancements made feasible by artificial intelligence (AI) and related technologies like the Internet of Things. The essay aims to demonstrate how AI and ML could be applied in this situation to improve the efficacy and accuracy of the public procurement system.

People analyze risks and task corrective actions to reduce the severity. But, in supply chain, risks cannot be eliminated, only reduced. Traditionally, supply chains involve three sub-sections, procurement, processing, and distribution. While processing is in-house; distribution and procurement are critical, and any issue there affects the whole supply chain. Therefore, resources are deployed to make them resilient and minimize impact. Procurement is defined as a process of buying goods and services. A procurement strategy needs to maintain a balance between the risks involved and create a win–win situation for both the organization and supplier. Since it involves decision-making at all levels and can affect every part of the organization, procurement strategy must prioritize risk mitigation to maximize its efficiency and efficacy. Hence, supply chain departments must realign their approaches to include evaluating suppliers, paying premium prices for raw materials, redefining product characteristics, identifying alternate suppliers globally, developing sources in the vicinity, changing logistics, outsourcing technology, and so on. This paper attempts to showcase how supply chain risks can be minimized using a structured approach, which includes supplier prioritization followed by risk management techniques, assessing impact, and creating a contingency plan to eliminate risks. A real case is presented in which the structured approach is discussed to mitigate risks in the nascent stage itself.

The importance and relevance of AI in current times cannot be over-emphasized. As with evolving technologies, AI too has grown in size and scope and permeated across disciplines and industry sectors. Starting from making the highly mundane and process-oriented tasks, into automation mode, it also provides a higher level of accuracy, comfort and economic value to both the industry applying AI and the consumer who gets a standard quality product or service at reasonable costs. AI and machine learning play thus a very crucial role in different processes and help improve accuracy, enhance the quality of prediction and enable better decisions. As always these involve rule-based heuristics processes, interactive processes, which are constructed to fit an algorithm, that can clearly express the relationship and process the data to give the final output. This has the aspect of higher accuracy, greater automation and lesser human intervention. Typically, the scenario that happens is that AI algorithms are applied using Machine learning applications, which then coordinate together to give the desired output. Typical areas where applications of AI can be seen are related to a Human Resource Management System (HRMS)––social recruiting, automated computerized job advertisements, candidate identification, flagging potential issues, apportionment of timeslots in line with the supervisor’s availability for job interviews as also automating the first level interview process. Over time it is expected that only selected positions will go for an interview by an HR manager and the rest may be done by the use of AI-ML integrated platforms, using bots. These have serious amplifications and consequences for the industry and how and what roles are likely to be taken up, depending on the convenience, payoff arising in the process, etc. This study examines the various aspects of these technologies and how they can help in the process of recruitment, accelerate the onboarding process and what factors help one to identify the process, examine the use of social media, social recruiting, virtual assessment, speeding up the hiring process and the like. Based on the specially designed questionnaire for this purpose, the authors propose an empirical study on a sample response of 281 individuals, across different levels from managers, senior executives and to examine how far the factors considered are responsible for explaining the impact of the changes happening in the application of AI and ML in the recruitment, selection and training process, and whether they have a significant role to play in the HRMS, and whether the use of AI and ML in HRMS is going to increase. Using factor analysis and regression analysis the authors studied the role played by the various factors and conclude that these factors have got a significant impact on the process of automating the HR process and application of various modern technologies including AI and ML in the system. The authors conclude that going forward, the HR process across industries is likely to be automated, with benefits in terms of quality, cost and efficiency flows accruing to the industry and final pass down to consumers in the form of reduced costs for products and services.

Many firms maintain highly complicated and worldwide supply chains in today’s world, which results in tons of transactions. Because of the gap in record maintenance for transactions, there arises the possibility of an anomalous transaction, i.e., any suspicious transaction that does not follow the historical pattern of regular transactions, thus creating a demand for disclosure of procurement and supply chain transactions. In this paper, the authors define the method of segregation of transactions and obtain derived parameters from a study. This also establishes a model-based correlation between the derived parameters and the blockchain. Blockchain plays a critical role in creating immutable records. It will help to reduce human intervention and induced errors, maintain transparency in the transactions, and improve the verifiability and traceability of an anomalous transaction during an audit.