Leveraging AI in SCM Literature Review

Introduction

This paper will discuss how artificial intelligence (AI) is being applied to the supply chain management (SCM) field. Specifically, this paper will cover how Toyota Motor Corporation (Toyota), a Global 500 company (Fortune, 2024), is implementing AI techniques to improve its SCM practices. This paper will cover Toyota, its supply chain, and its philosophy towards manufacturing in addition to the organization's approach to AI with respect to SCM, which strategies have been successful, which areas need improvement, and how their AI integration can be enhanced to optimize SCM.

Background

SCM has been defined in many ways. Some academics have taken a pragmatic approach to the definition, calling SCM a set of activities that involve “the flow of materials and products” (Habib, 2011, Section 2. Literature Review). Other academics have taken a higher-level approach, calling SCM a “management philosophy” (Habib, 2011, Section 2. Literature Review). In any case, the SCM field has several objectives, including, but not limited to, improving operational performance, outsourcing effectively, increasing profits, improving customer satisfaction, creating access to global markets, and addressing competitive pressures (Habib, 2011). Many SCM researchers conduct experiments to learn about how organizations can potentially add value, reduce costs, and slash response time with SCM techniques and principles (Habib, 2011).

AI has the potential to improve many SCM practices, including sourcing assessment/procurement, risk mitigation, inventory accuracy, logistics coordination, customer relations, and sustainability (Richey et al., 2023). However, organizations looking to implement AI into their SCM practices have several challenges they must contend with (Richey et al., 2023). These challenges include managing customer relations, training AI systems, operationalizing AI systems, managing AI biases, building cyber resiliency, and determining effective business models and use cases for AI systems (Richey et al., 2023).

Toyota Motor Corporation

Toyota is a major player in the global automotive industry (Toma & Naruo, 2017) and is currently ranked number 15 on the 2024 Fortune Global 500 list (Fortune, 2024). The company has more than 300,000 employees across all of its subsidiaries and “52 production facilities in 27 different countries” (Fujimoto et al., 2017, Section 2.1). This demonstrates the global nature of Toyota’s supply chain.

Toyota has a long history of leadership in manufacturing and SCM techniques (Toma & Naruo, 2017). This historic trend can be extrapolated to the current moment, where Toyota’s leadership can still be observed in areas such as the integration of AI techniques into its SCM practices (Dakshinamoorthy, 2022).

Toyota’s Supply Chain

SCM is a critical function in automotive companies due to the complexity that goes into manufacturing modern automobiles, which consist of approximately 20 to 30 thousand different components (Fujimoto & Heller, 2017). Another reason that SCM is so critical is that product and feature development often requires intensive collaboration with suppliers (Fujimoto et al., 2017). These supplier relationships are of critical importance due to the length of time that components must be supplied (Fujimoto et al., 2017). Toyota even goes as far as to view supplier factories as “an upstream extension of Toyota’s own production facilities” (Fujimoto et al., 2017, Section 2.2.3).

Toyota Production System

Toyota’s founder, Sakichi Toyoda, developed an approach towards manufacturing that ultimately evolved into the Toyota Production System (TPS) (Toma & Naruo, 2017), a production philosophy that has been studied across the world (Toyota Motor Corporation, n.d.). The three main pillars of the TPS are genchi genchi genbutsu, jidoka, and monozukuri (Toma & Naruo, 2017). Genshi genbutsu requires TPS adherents to go “to the source in order to solve problems and improve processes” (Toma & Naruo, 2017, p. 570). Jidoka is “autonomation/automation with a human touch” (Toma & Naruo, 2017, p. 570). Finally, monozukuri is “the spirit of manufacturing” (Toma & Naruo, 2017, p. 570).

These pillars of the TPS can be observed in how the company is approaching the integration of AI into its SCM practices. Specifically, genshi genbutsu can be interpreted as a data collection philosophy (Fujimoto et al., 2017). Researchers have claimed that the genshi genbutsu principle “suggests that data collection and decision making must be conducted in the place where the work is actually done” (Fujimoto et al., 2017, Section 2.2.3). This will be demonstrated later as Toyota’s implementation of real-time machinery monitoring technologies is studied (Dakshinamoorthy, 2022).

Demand Forecasting

Researchers have defined demand forecasting in SCM as “guessing future demand for components and vehicles based on [historical] sales data, market trends, and economic indicators” (Banerjee et al., 2024, VII. USE CASES). Toyota has integrated AI into its demand forecasting capabilities (Banerjee et al., 2024; Dakshinamoorthy, 2022). Toyota’s AI-integrated demand forecasting system allows it to dynamically predict changes in consumer preference (Dakshinamoorthy, 2022). This capability has allowed Toyota to “accurately [predict] demand for up to 6 months in advance, reducing the risk of stockouts or overstocking” (Banerjee et al., 2024, VII. USE CASES).

While forecasting itself may not be a novel concept to Toyota and similar businesses, AI techniques are providing massive improvements in accuracy. Many of Toyota’s historical forecasting techniques have been notably inaccurate (Xiaoying & Yan, 2010). This can be observed in the case of Toyota China, which historically struggled to produce accurate annual demand forecasts for its market (Xiaoying & Yan, 2010). In 2009, for example, Toyota China had a realized demand for a particular SUV model that was 90% higher than its forecasted demand (Xiaoying & Yan, 2010). For Toyota China, mismatches such as this one may not be easily resolved due to its dependence on Japan alongside several other countries for parts (Xiaoying & Yan, 2010). These downstream dependencies underscore the critical nature of accurate forecasting in SCM. Sourcing efforts are often dependent on demand forecasts, and cycle times at various stages can be delicately intertwined (Kordic, 2008).

Toyota is not alone in its efforts to apply AI techniques to demand forecasting; Shrivastav (2022) identified that successful usage of AI for demand forecasting has been seen throughout the literature alongside other aspects of strategy and operations planning. Shrivastav (2022) claims that organizations should expect some of the following results in response to leveraging AI for their demand forecasting efforts: accelerated business growth, improved experiences for customers, and improved efficiencies across the supply chain.

Predictive Maintenance

In addition to leveraging AI techniques for demand forecasting, Toyota is using AI for predictive maintenance, allowing Toyota to pre-emptively address potential equipment failures (Dakshinamoorthy, 2022). These efforts have allowed Toyota to reduce its maintenance costs and also improve its operational efficiency through a reduction in equipment downtime (Dakshinamoorthy, 2022). It’s important to note that while AI is an important enabling component of Toyota’s predictive maintenance efforts, another enabling component was the organization's implementation of real-time machinery monitoring technologies (Dakshinamoorthy, 2022).

Other organizations are using “various technologies such as vibration analysis, thermal imaging, and oil analysis” (Meddaoui et al., 2023, p. 3686) to enhance their predictive maintenance efforts. Data collected from these sensors are being fed into various machine learning algorithms, such as neural networks and random forests, to improve the accuracy of their predictions (Meddaoui et al., 2023).

Lessons Learned

One lesson that researchers took away from Toyota’s experience of implementing AI into its SCM practices was that AI must be applied systemically (Dakshinamoorthy, 2022). This systemic implementation is said to require pre-planning and organized execution of that plan in order to be effective (Dakshinamoorthy, 2022).

Another lesson that Dakshinamoorthy (2022) took away from Toyota’s experience is that there is value in using historical data points to inform future decisions. When this data is used to inform AI models, the value can be enhanced (Dakshinamoorthy, 2022).

Conclusion

Toyota was able to achieve operational improvements in its SCM practices through the use of AI techniques. Specifically, Toyota was able to improve its forecasting efforts, reduce the risk of over/understocking, reduce machine downtime, and reduce maintenance costs (Banerjee et al., 2024; Dakshinamoorthy, 2022). These efforts parallel practices that are being deployed in various SCM contexts (Meddaoui et al., 2023). Toyota has historically had difficulty producing reliably accurate demand forecasts (Xiaoying & Yan, 2010); however, new AI-enabled approaches have allowed them to achieve success with measurable impacts across the organization (Banerjee et al., 2024; Dakshinamoorthy, 2022).

Dakshinamoorthy (2022) identifies several challenges companies such as Toyota must contend with as they work to implement AI technologies into their SCM practices. These challenges include high upfront investment costs, data security concerns, and organizational friction as those companies implement changes (Dakshinamoorthy, 2022). However, Dakshinamoorthy (2022) has several recommendations for those companies.

First, Dakshinamoorthy (2022) recommends that organizations invest in AI infrastructure that will provide efficiencies as implementation efforts scale. Second, Dakshinamoorthy (2022) recommends that these organizations develop robust data governance frameworks to mitigate data security concerns. Third, Dakshinamoorthy (2022) recommends that organizations promote a culture of innovation, learning, and change to reduce internal resistance to new policies. Finally, Dakshinamoorthy (2022) recommends that organizations collaborate and form partnerships with AI vendors, other companies working on similar integration goals, and academia. This provides the benefit of information sharing across organizations with similar goals (Dakshinamoorthy, 2022)

Literature Gaps

Researchers have identified that “extensive studies on the role of AI in SCM remain relatively scarce due to a limited understanding of this phenomenon” (Helo & Hao, 2021, Introduction). Additionally, there is a noticeable lack of comprehensive case studies on how Global 500 companies are implementing AI into their SCM practices. Many case studies that do exist are operating on limited data points that inhibit the effectiveness of their research as well as the opportunities for subsequent researchers to conduct follow-up experiments. Dakshinamoorthy (2022), for example, conducts a case study on how Toyota and Unilever are applying AI to their SCM practices. However, they do not include citations for specific claims nor the raw data used to come to various conclusions.

Opportunities for Future Study

These gaps in the literature present opportunities for researchers to study the role that AI is playing in SCM as well as the opportunity for researchers to comprehensively evaluate how large organizations are applying AI at every stage of SCM. There is also an opportunity for a comprehensive case study to be performed on Toyota itself. Many departments across the organization are likely experimenting with and applying AI to various problems within its SCM practices. If this isn’t the case, there is a strong opportunity for researchers to study what the barriers to AI implementation in Toyota’s SCM practices currently are. Researchers have studied the barriers that exist on a more macroscopic level (Shrivastav, 2022); however, individual case studies on how Global 500 companies are integrating AI into their SCM practices could provide further insights.

References

Banerjee, A., Pawar, D., Kalambe, M., Jadhav, P., & Shukla, M. (2024). ARTIFICIAL INTELLIGENCE IN SUPPLY CHAIN MANAGEMENT FOR AUTOMOBILE INDUSTRY. Industrial Engineering Journal, 53(3), 1. https://www.researchgate.net/profile/Aditi-Banerjee-12/publication/382917939_ARTIFICIAL_INTELLIGENCE_IN_SUPPLY_CHAIN_MANAGEMENT_FOR_AUTOMOBILE_INDUSTRY/links/66b326f92361f42f23b919ce/ARTIFICIAL-INTELLIGENCE-IN-SUPPLY-CHAIN-MANAGEMENT-FOR-AUTOMOBILE-INDUSTRY.pdf

Dakshinamoorthy, D. (2022). Impact of Artificial intelligence (AI) on supply chain optimization. INTERNATIONAL JOURNAL OF ADVANCED RESEARCH IN ENGINEERING & TECHNOLOGY, 13(1). https://doi.org/10.34218/ijaret.13.1.2022.006

Fortune. (2024, August 4). Fortune Global 500. Retrieved November 15, 2024, from https://fortune.com/ranking/global500/?Name=toyota

Fujimoto, T., Marinov, M., Kato, Y., & Iwao, S. (2017). Strengthening Purchasing and Supply Chain Management at Toyota: A Lecture by Hiroaki Koda. In Industries and Disasters: Building Robust and Competitive Supply Chains. Nova Science Publishers, Incorporated.

Habib, M. (2011). Supply Chain Management : Applications and Simulations (M. Habib, Ed.). IntechOpen.

Helo, P., & Hao, Y. (2021). Artificial intelligence in operations management and supply chain management: an exploratory case study. Production Planning & Control, 33(16), 1573–1590. https://doi.org/10.1080/09537287.2021.1882690

Kordic, V. (2008). Supply Chain (V. Kordic, Ed.). IntechOpen.

Meddaoui, A., Hain, M., & Hachmoud, A. (2023). The benefits of predictive maintenance in manufacturing excellence: a case study to establish reliable methods for predicting failures. International Journal of Advanced Manufacturing Technology, 128(7–8), 3685–3690. https://doi.org/10.1007/s00170-023-12086-6

Richey, R. G., Chowdhury, S., Davis‐Sramek, B., Giannakis, M., & Dwivedi, Y. K. (2023). Artificial intelligence in logistics and supply chain management: A primer and roadmap for research. Journal of Business Logistics, 44(4), 532–549. https://doi.org/10.1111/jbl.12364

Shrivastav, M. (2022). Barriers related to AI implementation in supply chain management. Journal of Global Information Management, 30(8), 1–19. https://doi.org/10.4018/jgim.296725

Toma, S. G., & Naruo, S. (2017). Total Quality Management and Business Excellence: The Best Practices at Toyota Motor Corporation. Amfiteatru Economic, 19(45), 566–580.

Toyota Motor Corporation. (n.d.). Toyota Production System. Toyota Motor Corporation Official Global Website. Retrieved November 16, 2024, from https://global.toyota/en/company/vision-and-philosophy/production-system/

Xiaoying Liang, & Houmin Yan. (2010). Inventory models with delivery upgrade. 2010 7th International Conference on Service Systems and Service Management, 1–6. https://doi.org/10.1109/ICSSSM.2010.5530246