A systematic review of machine learning in logistics and supply chain management: current trends and future directions

Increasing load factor in logistics and evaluating shipment performance with machine learning methods: A case from the automotive industry

The insufficient loading of vehicles, which leads to a low logistics load factor is a common problem in the logistics industry. This study addresses this issue by utilizing actual shipment data from an automotive company. An effective method has been proposed to improve the company's logistics efficiency through a scenario-based approach. Two real- world scenarios were developed to enhance vehicle loading performance. Machine learning algorithms were employed to evaluate the shipment performance of these scenarios. For the study, a dataset was generated from the company's ERP system and real-world shipment data. Since this is a real-world problem, the dataset consisted of unlabeled data. To solve this problem, both supervised and unsupervised learning algorithms were applied. First, unsupervised clustering algorithms were used to group the shipment performance based on similarities. Then, supervised learning algorithms were utilized to classify the data within each group. The average cost was used to evaluate the clusters obtained through the unsupervised methods, while classification performance was measured using supervised machine learning techniques. The scenario-based approach has significantly improved the performance of the shipments as it shows the changes in load factor more clearly. In the actual case, only %25.7 of shipments were high performing, while this percentage gradually increased to %98.4 in the scenarios. The results show that optimizing the load factor makes the transports more efficient and balanced.

Managing from a distance in international purchasing and supply

International purchasing and supply management (PSM) teams have long faced the visibility and understandability challenges of managing geographically dispersed and culturally distant suppliers. Problems arising from inadequate monitoring and control over suppliers can be attributed to geographical and cultural distance, capability gaps, weak institutions, and supply market dynamism. With transaction costs theory as our lens, we examine how international geographically and culturally distant purchasing and supply management (PSM) teams control emerging economy suppliers with formal management controls. We use interview survey data on 339 international customer-Chinese supplier relationships using supplier perceptions of the extent to which performance measurement and monitoring practices are used by their primary customer in the purchase reorder decision and control. The results demonstrate that the cultural and, to a lesser extent, geographical distance between the customer and the supplier is associated with more extensive use of formal management controls. Also, we find the relationship between geographical or cultural distance and the importance of performance measurement is strengthened for suppliers of complex components.

Modelling the relationships between the barriers to implementing machine learning for accident analysis: the Indian petroleum industry

Purpose

This paper aims to identify, prioritise and explore the relationships between the various barriers that are hindering the machine learning (ML) adaptation for analysing accident data information in the Indian petroleum industry.

Design/methodology/approach

The preferred reporting items for systematic reviews and meta-analysis (PRISMA) is initially used to identify key barriers as reported in extant literature. The decision-making trial and evaluation laboratory (DEMATEL) technique is then used to discover the interrelationships between the barriers, which are then prioritised, based on three criteria (time, cost and relative importance) using complex proportional assessment (COPRAS) and multi-objective optimisation method by ratio analysis (MOORA). The Delphi method is used to obtain and analyse data from 10 petroleum experts who work at various petroleum facilities in India.

Findings

The findings provide practical insights for management and accident data analysts to use ML techniques when analysing large amounts of data. The analysis of barriers will help organisations focus resources on the most significant obstacles to overcome barriers to adopt ML as the primary tool for accident data analysis, which can save time, money and enable the exploration of valuable insights from the data.

Originality/value

This is the first study to use a hybrid three-phase methodology and consult with domain experts in the petroleum industry to rank and analyse the relationship between these barriers.

Current and Future Trends of Information Technology and Sustainability in Logistics Outsourcing

Sustainability awareness across various economic sectors requires firms to use logistics outsourcing to reduce logistics-related emissions with compliant logistics service providers (LSPs). LSPs apply advanced information technologies to help achieve high efficiency, effectiveness, and sustainability goals. While logistics outsourcing has received considerable attention from researchers, limited research has identified the elements of logistics outsourcing and established research trends regarding technology and sustainability aspects of logistics outsourcing. This study aims to establish trends in technology and sustainability in logistics outsourcing and identify the important elements of logistics outsourcing. A bibliometric analysis technique using the Biblioshiny package in R. A SCOPUS search resulted in 2019 documents. Research on technology and sustainability in logistics outsourcing is growing, especially in developed countries, with little or no research from developing countries. Sustainable supply chains and third-party logistics themes dominated the past research. Current research is on reverse logistics, circular economy, and green logistics; the latter is likely to dominate the future, focusing on risk management, pollution control, and innovation through advanced technologies such as artificial intelligence, machine learning, and big data. Essential elements of logistics outsourcing are identified as maintaining a good relationship based on trust and cooperation, LSP multi-selection criteria that include sustainability and technology capabilities, proper contract management, and an appropriate in-house versus outsourcing balance for competitiveness. LSP managers are advised to develop sustainability and technology capabilities, including reverse logistics. The paper contributes to logistics management theory by identifying the elements of logistics outsourcing and presenting a bibliometric result to guide future research on sustainability and technology capabilities in logistics outsourcing.

Digital technologies as enablers of supply chain sustainability in an emerging economy

Vietnam is a country with significant potential for growth as a global centre for manufacturing, as supply chains look to reduce their over-reliance on China in the aftermath of COVID-19. The objective of this study is to better understand the current adoption rates and growth potential of emerging Industry 4.0 (I4.0) digital technologies and ascertain their potential to drive successful future sustainability initiatives amongst Vietnamese supply chain firms. These technologies offer a wide range of sustainability benefits, from a potential to reduce waste production and lower energy consumption to increased opportunities for recycling and industrial symbiosis. This empirical study surveys 223 Vietnamese supply chain experts to learn how digital technologies are being utilized in that region, what levels of future investment are expected, what preparatory measures are being taken to leverage new technologies, and what scope for improved supply chain sustainability exists. The findings indicate a low level of I4.0 digital technology adoption amongst Vietnamese supply chain firms, with the Internet of Things (IoT) currently being the most prevalent (48 percent adoption rate). Drones, Big Data Analytics and IoT are the I4.0 digital technologies expected to have the greatest future impact on Vietnamese supply chains. Whilst I4.0 digital technology adoption is still at this early stage, that may present a greater opportunity for driving future sustainability outcomes, than interrupting and retrofitting solutions to already-established networks and infrastructure.

Last mile delivery in logistics and supply chain management: a bibliometric analysis and future directions

Purpose

The main objective of this paper is to provide a systematic literature review (SLR) and structured insight into last mile delivery, ultimately identifying gaps in current knowledge and proposing a framework for future research direction in terms of sustainability in the area.

Design/methodology/approach

This paper identifies and synthesizes information from academic journals and examines “Journals and Publishing place,” “Geographic location,” “Year of Publication,” “University and Author Affiliation,” “Themes and Sub-themes,” “Theory,” “Research Design, Methods and Area” and “Industry Involvement.” A collection of online databases from 2005 to 2020 were explored, using the keywords “Last mile delivery,” “Last mile logistics,” “Last mile transportation,” “Last mile fulfillment,” “Last mile operations” and “Last mile distribution” in their title and/or abstract and/or keywords. Accordingly, a total of 281 journal articles were found in this discipline area, and data were derived from a succession of variables.

Findings

There has been significant growth in published articles concerning last mile delivery over the last 15 years (2005–2020). An in-depth review of the literature shows five dimensions of the last mile: last mile delivery, transportation, operations, distribution and logistics. Each of these dimensions is interrelated and possess clustered characteristics. For instance, last mile operations, last mile transportation and last mile delivery are operational, whereas last mile distribution is tactical, and last mile logistics possess strategic characteristics. The findings also indicate that even though the sustainability concept can be incorporated into all levels of the last mile, the current literature landscape mainly concentrates on the operational level.

Research limitations/implications

This review is limited to academic sources available from Emerald Insight, Science Direct, Taylor and Francis, Springer, MDPI and IEEE containing the mentioned keywords in the title and/or abstract/or keywords. Furthermore, only papers from high-quality, peer-reviewed journals were evaluated. Other sources such as books and conference papers were not included.

Practical implications

This study dissects last mile delivery to produce a framework that captures and presents its complex characteristics and its interconnectedness with various related components. By analyzing last mile delivery in its entirety, the framework also helps practitioners pinpoint which levels of last mile delivery (operation, tactical or strategic) they can incorporate the concept of sustainability.

Originality/value

The research findings enrich the contemporary literature landscape and future work by providing a conceptual framework that incorporates the “economic,” “environmental” and “social” pillars of sustainability in all dimensions of the last mile delivery.

Does COVID-19 influence the maturity management of industrial sectors? Evidence from global data

Purpose

This study aims to understand how the COVID-19 pandemic dramatically impacts the maturity of all industrial sectors globally. This paper analyses the general patterns of managing maturity in terms of performance and risk-taking of S&P 500 industrial sectors while determining their association with COVID-19.

Design/methodology/approach

To analyse the immediate response of COVID-19 on maturity management, the authors gather time-series daily index data of S&P sectors from October 2019 until June 2020 from Bloomberg. The authors select this study period to show the immediate effect of COVID-19 on industrial sector maturity management. The performance and volatility of stock are proxies for managing the maturity of each sector. The authors use vector auto-regression (VAR) methodology to determine the impact of global coronavirus.

Findings

This study’s findings suggest that the information technology sectors outperform the other sectors; in contrast, the utility sector exhibits the worst performance during a pandemic. Furthermore, the real estate sector depicts a higher level of systematic risk pattern than other sectors. Interestingly, the empirical result of VAR shows that almost every sector is significantly negatively affected by this pandemic; however, the consumer discretionary sector is immune to it.

Research limitations/implications

Overall, this study’s findings for individual economic sectors demonstrate that the managing maturity of each sector acts differently to the coronavirus outbreak. This study offers insights to researchers, policymakers, regulators, financial report users, investors, employees, clients and society.

Originality/value

This paper contributes to the existing literature on managing the maturity of industry sectors in terms of observing their trends during the financial crisis.

Illustrating scholar–practitioner collaboration for data-driven decision-making in the optimization of logistics facility location and implications for increasing the adoption of AR and VR practices

Purpose

This study illustrates the experience of scholar–practitioner collaboration for data-driven decision-making through the problematic of optimizing facility locations and minimizing logistics costs for La Palette Rouge (LPR) of Portugal.

Design/methodology/approach

The authors used a mixed mixed-method approach involving (1) a quantitative exploratory analysis of big data, which applied analytics and mathematical modeling to optimize LPR's logistics network, and (2) an illustrative case of scholar–practitioner collaboration for data-driven decision-making.

Findings

The quantitative analysis compared more than 20 million possible configurations and proposed the optimal logistics structures. The proposed optimization model minimizes the logistics costs by 22%. Another optimal configuration revealed that LPR can minimize logistics costs by 12% through closing one of its facilities. The illustrative description demonstrates that well-established resource-rich multinational enterprises do not necessarily have the in-house capabilities and competencies to handle and analyze big data.

Practical implications

The mathematical modeling for optimizing logistics networks demonstrates that outcomes are readily actionable for practitioners and can be extended to other country and industry contexts with logistics operations. The case illustrates that synergistic relationships can be created, and the opportunities exist between scholars and practitioners in the field of Logistics 4.0 and that scientific researcher is necessary for solving problems and issues that arise in practice while advancing knowledge.

Originality/value

The study illustrates that several Logistics 4.0 challenges highlighted in the literature can be collectively addressed through scholar–practitioner collaborations. The authors discuss the implications of such collaborations for adopting virtual and augmented reality (AR) technologies and to develop the capabilities for maximizing their benefits in mature low-medium technology industries, such as the food logistics industry.

Applications of artificial intelligence and machine learning within supply chains:systematic review and future research directions

Purpose

The purpose of this study is to investigate how artificial intelligence (AI), as well as machine learning (ML) techniques, are being applied and implemented within supply chains (SC) and to develop future research directions from thereof.

Design/methodology/approach

Using a systematic literature review methodology, this study analyzes the publications available on Web of Science, Scopus and Google Scholar that linked both AI and supply chain from one side and ML and supply chain from another side. A total of 388 research studies have been identified through the before said three database searches which are further screened, sorted and finalized with 50 studies. The research thoroughly reviews and analyzes the final lists of 50 studies that were found relevant and significant to the theme of AI and ML in supply chain management (SCM).

Findings

AI and ML applications are still at the infant stage and the opportunity for them to elevate supply chain performance is very promising. Some researchers developed AI and ML-related models which were tested and proved to be effective in optimizing SC, and therefore, the application of AI and ML in supply chain networks creates competitive advantages for firms. Other researchers claim that AI and ML are both currently adding value while many other researchers believe that they are still not fully exploited and their tools and techniques can leverage the supply chain’s total value. The research found that adoption of AI and ML have the ability to reduce the bullwhip effect, and therefore, further supports the performance of supply chain efficiency and responsiveness.

Research limitations/implications

This research was limited in terms of scope as it covered AI and ML applications in the supply chain while there are other dimensions that could be investigated such as big data and robotics but it was found too lengthy to include these additional dimensions, and therefore, left for future research studies that other researchers could explore and pursue.

Practical implications

This study opens the door wide for other researchers to explore how AI and ML can be adopted in SCM and what are the models that are already tested and proven to be viable. In addition, the paper also identified a group of research studies that confirmed the unexploited avenues of AI and ML which could be of high interest to other researchers to explore.

Originality/value

Although few earlier research studies touch based on the AI applications within manufacturing and transportation, this study is different and makes a unique contribution by offering a holistic view on the AI and ML implications within SC as a whole. The research carefully reviews a number of highly cited papers classifying them into three main themes and recommends future direction.