Rethinking Supply Chain Performance Metrics in the Artificial Intelligence and Digital Age

Rethinking Supply Chain Performance Metrics in the Artificial Intelligence and Digital Age

Nubi Achebo, PhD

Nigerian University of Technology and Management Apapa-Lagos

DOI: https://dx.doi.org/10.47772/IJRISS.2025.909000049

Received: 30 August 2025; Accepted: 05 September 2025; Published: 28 September 2025

ABSTRACT

Deep within global supply chains, performance is being transformed as artificial intelligence (AI) and digital technology redefine what is possible and how these new possibilities can be measured. Old-school measures like cost efficiency, customer service levels and inventory turns can be helpful, but increasingly do a poor job reflecting the agility, resilience, customer centricity and sustainability that is the hallmark of success in today’s hyper-volatile and highly networked economy. This article challenges the limits posed by traditional metrics and re-imagines the horizons of supply chain measurement by investigating the transformative powers that AI-enabled analytics, automation, predictive modeling and real-time data processing hold for re-shaping  performance. It explores how digitalization offers more agile, transparent and customer centric metrics with examples of case studies from various sectors and reviews evolving practices as well as the cultural, technological and data privacy challenges faced by practitioners. The paper posits that rethinking supply chain metrics is not just a matter of operational requirement; rather, it is a strategic imperative for firms desiring long termcompetitive advantage, robustness and alignment with global sustainability objectives.

Keyword: Supply Chain, Supply Chain Metrics, Performance Metrics, Artificial Intelligence, Digitilization

INTRODUCTION

Supply chains are among the most intricate and critical lifelines linking organizations worldwide, given the large number of moving parts involved in planning, sourcing, production, warehousing, distribution, and delivery. Metrics used to assess their efficient management have traditionally focused on cost, service, and flexibility, but artificial intelligence (AI) and digital technologies now create opportunities for more useful measures that better capture value.

Supply chain management (SCM) has its origins in the physical distribution management concepts introduced in the 1950s and developed further through the 1960s and 1970s, albeit mainly in distribution-oriented companies rather than manufacturers. The term “supply chain” first appeared in the literature in 1982 and evolved quickly during the 1980s, culminating in the establishment of a journal in1989. By the 1990s the supply chain concept had seized the imaginations of business leaders and the topic was firmly positioned on the academic agenda. The influence of digital technologies emerged in the late-1990s, becoming a major topic of interest in the early-2000s.

Traditional supply chain metrics established during its earlier evolution remain relevant, describing aspects such as cost efficiency, customer service levels, and inventory turnover. A focus on cost efficiency underscores the importance of keeping supply chain operational expenses at a minimum while still meeting demand. Service-level metrics gauge the effectiveness in fulfilling customer orders promptly and accurately, reflecting the supply chain’s responsiveness. Inventory turnover rates indicate how efficiently inventory is managed, with higher turnover suggesting a lean and agile supply process. Yet the growing complexity accompanying the emergence of new digital technologies, along with the increasing paucity of raw materials, implies that evolving these conventional criteria will be necessary going forward. More broadly, the prevalent focus should be on realizing the full potential of digital technologies and artificial intelligence. Efficiency alone provides only very limited guidance for the types of supply chain performance that should be aimed for when embracing the opportunities offered by these technologies (Zapke, 2019, Perano, et al., 2023).

Supply Chain Management Evolution

Supply chain operations have progressed through several evolutionary levels. Beginning with rudimentary integration, the field advanced to functional integration which involved creating materials management and physical distribution functions and departments. In the following level of internal integration, manufacturers worked with suppliers and customers to streamline the flow of materials and information. This was succeeded by external integration typified by such initiatives as electronics data interchange (EDI). The most recent stage involves digital integration, an emerging adaptation to new eras defined by digital technology, platforms, and ecosystems. This progression illustrates how manufacturing firms have evolved from managing their supply chains with labor-intensive processes to adopting digital and automated methods (Rana & Daultani, 2022). It parallels the transformation from traditional, labor-intensive operational supply chains to digital supply chains driven by industry 4.0 technologies .

Traditional Performance Metrics

Supply chain output is traditionally measured by cost, efficiency, and service levels. Information quality or decision accuracy is equally important, especially when AI systems support logistics (Zapke, 2019). Environmental considerations such as carbon emissions and sustainability are becoming common corporate priorities. Emerging technologies such as 3D printing, nanotechnology, intelligent agents, and distributed networks facilitate decentralization, automation, integration, and enhanced supply chain responsiveness.

Cost Efficiency

Cost efficiency remains a primary criterion in many organizations. Metrics often include inventory turnover rates and warehouse costs, providing a baseline for evaluating resource utilization and operational effectiveness. These measures offer a simple, repeatable, and objective way to benchmark performance. Improvement initiatives generally correlate with a cost-effective supply chain.

Service Levels

Service levels represent the proportion of customer demand met without delay. Objectives might target a “Five 9s” level, implying uninterrupted delivery to customers. High service levels translate into superior customer service and fewer costly order problems and expediting efforts. Deviations often incur overtime, subcontracting, and rush transport, underscoring the financial impact of shortfalls. Accurate supply chains encounter minimal penalties and decreased customer dissatisfaction.

Inventory Turnover

Inventory turnover—the number of times inventory cycles through annual sales—is a broad measure linked to both cost and service. Operational enhancements often affect inventory turnover by reducing cycle stock or improving using buffer stock. In stable environments, better forecasting reduces surprises, diminishing the need for dead stock. Hence, turnover reflects supply chain efficiency, with implications for cash flow and capital employed.

These traditional metrics become insufficient in dynamic, digitally integrated supply chains, highlighting the need for metrics that address agility, quantifiable customer happiness, and sustainability.

Cost Efficiency Metrics

Reducing costs remains the prime consideration in supply chain management. Performance evaluation traditionally hinges on three critical metrics: cost efficiency, service levels, and inventory turnover (Zapke, 2019). Most enterprises are primarily cost-driven, and supply chains typically represent one of the largest cost components. Hence, cost efficiency serves as the foremost indicator.

Organizations develop optimal sourcing strategies based on the costs of alternative suppliers. A prudent sourcing strategy balances cost with available support services, acknowledging that the cheapest option might not invariably deliver the best outcome. Cost analysis informs decisions around production plants
location and ownership and dictates the choice between crafting goods internally or outsourcing. Virtual Enterprises provide greater flexibility, enabling production in proximity to specified markets, sometimes closer than traditional suppliers or in distinct regions to exploit differential labor rates (Abdu Alomar, 2022).

Service Level Metrics

Service levels indicate the capability to fulfill customer demand within a specified timeframe and remain paramount in contemporary supply chain management strategies. The service-level metric itself denotes the proportion of orders or order lines delivered in full and on time. Supporting metrics such as perfect orders, fill rate, order line fill rate, order cycle time, and order inaccuracy serve as sub-indicators of this primary measure. Service-level considerations extend beyond logistics to encompass planning lead times, effective inventory positioning, and even order acceptance protocols.

Inventory Turnover Metrics

Inventory turnover measures how well a firm manages intermediate inventories. It is calculated by dividing the amount of raw materials or goods processed within one period by the average inventory during the same interval. Values should exceed or approximate 1. Due to variations in demand, production rates, and local sales, product units cannot be transported consistently, increasing MOQ (minimum order quantity) and consequently the average inventory level. In line with this, inventory turnover represents the number of times the average inventory is sold annually. Increasing inventory turnover and days of supply can elevate management and control efforts and may boost costs (Olaniyi & Pugal, 2024).

Limitations of Traditional Metrics

Supply chain performance metrics have continued to represent a widely discussed question where key performance indicators cover a wide choice that reflect the current industry status and situation. Despite a wide coverage of metrics including cost, agility, quality, time and value-based metrics , the choice is mostly related to the specific industrial segment and the available technologies. The rise of AI and the digital supply chain might significantly alter the existing patterns towards the adoption of more AI-related services and business solutions. Technological integration and complexity drive Supply Chain Management (SCM) towards a greater adoption of systems with the means to effectively harness these solutions and manage the overall network (Abdu Alomar, 2022) ; (Zapke, 2019). However, the rapidly evolving and experimental technological characteristics, as well as the lack of a sustainable framework, render any conclusion precarious without the support of preliminary data and evidence. Multi-disciplinary approaches aim to ensure acceptance through awareness and knowledge dissemination and to facilitate the adoption of integrated solutions . The prevalent employment of metrics reflecting the traditional efficiency paradigm presented by cost, time and quality-based indices, which prioritise limitation of costs and maximum utilisation of capabilities without fully reflecting the sustainability profile, emphasises the significance of Metrics for the Digital Supply Chain (MSDC) beyond efficiency as a fundamental topic that remains open and requiring consideration in view of the multifaceted potential outcomes in the near future.

Establishing the current industrial alignment of the paradigm demands emphasis on the prevalent metrics that are primarily cost or operationally related even within recent work , without effectively addressing the responsibility and sustainable design underpinning the Digital Economy philosophy. This approach leverages an extensive five-year-updated dataset to present a thorough single-industry network assessment that fundamentally addresses such shortcomings, highlighting the direction in which the paradigm is swiftly progressing.

Inflexibility in Dynamic Markets

The increasing turbulence in the marketplace requires supply chains to manufacture products and take strategic and operational decisions under uncertainty (Chatterjee & Chaudhuri, 2022).  Since supply chains must now operate in increasingly volatile, uncertain, complex, and ambiguous environments, decision-making may well have to occur with incomplete, unreliable, or outdated information. Consequently, supply chain performance metrics must adapt to these new conditions and be redefined to meet the emerging requirements of the competitive landscape.

Short-term Focus

Traditional performance metrics tend to focus on short-term outcomes. Low costs, high service levels, and rapid inventory turnover are all preferable, but this gives supply managers little chance to recover from major shortfalls. Companies unable to trade flexibly with a margin for error must underinvest, deferring projects that may compromise near-term productivity. Even though the horizon inevitably shifts, short-term metrics are especially ill-suited to a global sector undergoing rapid structural change; response lags are too long, and the path from investment to results too uncertain. Once again, the implication is that firms increasingly able to trade with confidence should be encouraged to invest heavily, expanding the horizon still further (Abdu Alomar, 2022).

Neglect of Customer Experience

Transportation systems are optimised based on prices, leading to economic savings without additional workload. However, this efficiency emerges from two contributing factors: the presence of a flexible agent capable of negotiating transportation costs, and the absence of greater communication complexity. Alongside cost-efficiency, customer satisfaction, and sustainability, agility defines the ability of an organisation or a system to react to disruptive changes in the medium-to-long term (Abdu Alomar, 2022) ; (Zapke, 2019).

Adaptive methods such as deep learning neural networks provide an effective means to assess supply chain performance grades and project future trends. Utilising multi-layer and bi-directional structures, these contemporary approaches adeptly resolve longstanding challenges like vanishing and exploding gradients. By projecting estimated values into discrete observation classes, these networks synthesise missing data and refine consideration of multiscalar relationships and supply chain evolution dynamics.

Following the crucial insights of Abdu Alomar, this exploration advocates for a reconsideration of supply chain performance metrics in light of technologies in AI and digital supply chain.

AI in Supply Chain

Artificial intelligence (AI) is transforming supply chains by enabling the adoption of metrics that extend beyond the traditional focus on efficiency. As supply chains and their surrounding business environments have become more digital, AI systems have emerged as a central element of supply chain digitalisation. Several years of studies and real-world experiments have identified AI also as a key enabler for the deployment of novel and more relevant performance metrics (Zapke, 2019).

The industrial and information revolutions prompted iterative phases of adoption, which gradually transformed supply chains and their management. Driven by these transformations, metrics widely accepted in the discipline appeared insufficient to guide decisions in disturbed and flexible contexts. For instance, disruption in the supply chain as a result of COVID-19 and Russia-Ukraine conflicts dramatically exposed the weakness of a cost-based approach. Supply chain disruption revealed the need to rethink criteria used to define performance in the digital age and prompt managers to address a broader scope 9 including resilience, sustainability and the customer experience 9 as a whole.

The process industries inevitably initiated a paradigm shift focused on supply chain performance metrics. Further advances remained limited, as electrification  merely contributed to extending the classical models into peripheral areas of supply chains such as warehousing. At the turn of the 21st century, digitalisation sparked renewed momentum 9 beyond digital technologies like ERP systems. A gradual transition from technology-centred to supply chaincentric management (Industry 4.0) has unfolded. Raw data are processed and analysed at scale with an increasing focus on real-time conditions. Based on these advances, updated business models emerge: Market4.0, Industry 5.0 or Supply Chain 5.0. They are characterised for example by a shift from ownership to stewardshipand, more generally, from linear to quantum and digital transformations.

AI technologies and digitalisation fundamentally impact supply chains as well as their performance evaluation. They are reshaping operations with the adoption of advanced automation and analytic capabilities, which enables the processing of fragmented and complex demands through new governance schemes.

Predictive Analytics

Predictive analytics has become an integral element of business intelligence due to increasing operational complexity, rising volumes of available business data, the need to reduce cost, the imperative to grow revenue, and the rising risk exposure (Stefanovic, 2014). Incorporating predictive analytics into inventory management can improve financial performance by lowering costs and simultaneously preserving customer service, enabling optimal reordering policies, raising productivity, reducing the cash-to-cash cycle time, and increasing profitability. Data mining techniques including neural networks and autonomous agents have been extensively applied to optimize inventory levels and to forecast product prices. When integrated with reporting and analysis, as well as data discovery, data quality, and data integration, predictive analytics starts to affect all steps of the analytic process and enables careful monitoring of business status.

An infrastructure ready to perform predictive analysis at all stages of data management enhances supply chain agility and strengthens the competitive advantage of any enterprise. Predictive analytics are expected to become a central component of supply chain performance management (SCPM) systems, which measure and compare strategic metrics to ensure optimal use of resources. When combined with historical data and current business status, it becomes possible to predict key performance indicators such as cost, orders fulfilment, and Return On Working Capital (ROWC). Built on such forecasts, SCPM systems will allow on-the-fly monitoring of the entire supply chain and enable timely inventory replenishment decisions as well as continuous adaptation at strategic, tactical, and operational levels, thereby further increasing business performance. McKinsey research estimated that predictive analytics could reduce inventory by 10 20 percent and increase service levels at the same time (Zapke, 2019). Automation and demand forecasting represent substantial and trustworthy types of predictive analytics.

Human-robot collaboration will also significantly alter manufacturing processes thus improving efficiency, safety, and product quality. In warehousing, predictive AI enhances decision-making, enabling inventory reductions of up to 30 percent, and allows larger portions of order fulfilment to be automated through cognitive robots.  E-commerce changed demand for immediate delivery, which AI can address through optimized routing based on historical and real-time traffic data, and by powering autonomous trucks and drones. Overall, AI is expected to transform manufacturing, warehousing, and distribution processes, boosting efficiency and reducing costs.

Era of Automation and Robotics

Automation and robotics generate data and decisions, replacing or augmenting human activities (Zapke, 2019). Automation is increasingly realistic in retail (Amazon Go), warehousing (Ocado Smart Platform), manufacturing (Siemens), and other functions. However, automation’s purpose differs from that of analytics: automation frequently targets operational efficiency, while business analytics targets strategic advantage. Automation can generate additional data that distinctive analytics models can extract first-mover business benefits from, and it often provides better milled input data for these models.

Advanced robots often demand less human support, and clearly can generate significant cost savings (Abdu Alomar, 2022). Automation also lowers sensitivity to human labour issues, such as strikes and health restrictions, and can work continuously without the fluctuations associated with shift systems. Automated manufacturing therefore has the potential to enable just-in-time (JIT) production and delivery, producing a very attractive increase in flexibility and agility. The associated increase in responsiveness also supports customers adapting product choices and delivery times at short notice. Moreover, physical supply chains commonly exhibit greater inertial delay than their digital data pathways. A switch to greater automation therefore offers potential to reduce supply chain friction and raise needed agility.

Real-time Data Processing

Rapid events can now be addressed quickly through the real-time processing of incoming information. The deployment of artificial-intelligence (AI) algorithms at scale within supply-chain networks allows tasks that once required dedicated specialists to be undertaken automatically. Digital solutions that exploit AI enable network orchestration to be optimized across nodal locations, thereby delivering improved efficiency and effectiveness (Abdu Alomar, 2022). Automation of interactions with customers through chatbots and voice assistants offers an especially attractive route to customer support that functions efficiently and increases customer satisfaction while generating an improved return on investment.

The 4th Industrial Revolution has highlighted a number of areas within the supply chain that are poised to benefit significantly from the application of AI. Operational procurement already features the use of smart data and chatbots, enabling purchases to be executed with unprecedented agility. Supply-chain planning is able to integrate AI data and analytics to forecast demand, leading to further benefits in warehouse management through improved inventory-optimization techniques that, in turn, have a direct impact on transportation and shipping by reducing the total costs of delivery (Nozari et al., 2022). Supplier selection increasingly incorporates the use of up-to-date data covering alternative options while having the ability to monitor continuously supplier performance through the sourcing process.

New Performance Metrics for the Digital Age

Traditionally some performance indicators such as costs, customer satisfaction, and on-time delivery—remain essential, data-driven approaches enable the creation of more meaningful metrics (Stefanovic, 2014).Three key metrics address the deficiencies of conventional KPIs—agility, customer-centric, and sustainability.Agility emphasizes flexibility, transparency, and velocity. Customer-centricity considers convenience, price, and experience. Sustainability measures include ecological footprint and environmental impact. Agility demonstrates the rapid responsiveness that digitally driven supply chains can achieve. Customer-centric metrics recognize that success is no longer determined by logistics performance alone, but by offering the speed and convenience customers seek. The need for economically and socially responsible operations—already a critical concern—grows stronger as awareness of climate change expands (Zapke, 2019).

Agility and Responsiveness

Supply chain agility—a firm’s capacity to meet fluctuations in demand and supply—offers an alternative to the low-cost model that prioritizes cost effectiveness (Gunasekaran et al., 2018). One study defines supply chain agility as proactivenesss in responding rapidly to unexpected changes in demand or supply. Digital supply chains can enhance this attribute using real-time data, collaboration software, and computer systems that enable coordination up and down the chain. Agility is especially valuable during periods of change, when a supply chain must quickly alter its course. Practical examples of companies excelling in this domain confirm the feasibility of maintaining agility at scale.

Responsiveness relates to shorter cycle times. The idea of responsiveness underpins the physical Internet standard, based on the use of rapid transfers and worldwide warehouse slots. One study demonstrates that cycle time correlates significantly and positively with responsiveness. Increases in cycle time reduce supply chain responsiveness; to be more responsive, organizations need it to be as short as possible. A key enabler of cycle-time reduction is digital connectivity, which sustains the faster physical flows required. Automated vehicles and digital tracking systems are just two examples among many. 

Customer-Centric Metrics

Customers expect seamless and frictionless multi-channel transaction experiences. Consequently, customercentric metrics gain increasing importance in the digitalised supply chain. Customer-centric key performance indicators (KPIs) focus on measuring the ability and capacity of the supply chain to satisfy customers’ purchasing and post-purchase behaviours and to meet customers’ activity expectations (Abdu Alomar, 2022). Measures including the Net Promoter Score (NPS), after-sales support performance, customer satisfaction, return requests, average cost per return, and average time for returns provide the basis for current customercentric kpis. Ensuring sustainability and supply chain resilience raises the importance of such performance metrics in the digital age (Zapke, 2019).

The digital transformation greatly accelerates the complexity of trade in internet of things (IoT) enabled supply chains. Great examples of transformation in digital enabled supply chains can be found in Amazon, Walmart, Avaya, Unilever and Pfizer.  AI enables the prediction of user behaviour and the tracking of product demand, and holds the key to autonomous supply chain management. Various researchers, including , and , have proposed maintaining the improved overview of a supply chain through comprehensive control towers and masterchains, considering the overarching category of flexibility in addition to the transportation-focused definition of flexibility suggested by . The study integrates insights from previous sections 2 “The Evolution of Supply Chain Management”, 3 “Traditional Performance Metrics”, 4 “Limitations of Traditional Metrics”, and 5 “The Role of AI in Supply Chain”.

Open research questions explore operationalisation and guidance for supply chain organisations to manage and measure these emerging requirements. As larger firms set new digital standards and observe increasingly digitalised supply chain operations, medium-sized enterprises face the challenge of a substantial paradigm shift in supply chain management. This shift impacts current strategies, structures, processes, and performance management, together shaping the future of the discipline. The realisation and operationalisation of flexible supply chains remain unanswered, as evidenced by previous investigations .

Sustainability Metrics

Sustainability is now a critical consideration managing supply chain. Managers are pressured to achieve sustainability objectives while also preserving traditional goals such as efficiency, responsiveness, service, and safety (Deng et al., 2024). The relevant literature demonstrates substantial efforts to develop sustainability-related metrics, yet the prevalence of sustainability dimensions across these work remains limited (Tuni & Rentizelas, 2019). Many models are strongly focused on environmental sustainability, considered in the context of economic aspects through cost and risk measurements, whilst the social dimension tends to be neglected (Acquaye et al., 2017). In developing an assessment methodology, several criteria enable a comprehensive sustainability evaluation that is not overly demanding in terms of data availability. Organisational awareness and commitment are required, but the ad hoc calculation of performance indices using large external databases is not necessary. The approach is also suitable for implementing continuous improvement processes, as it supports comparison between current and previous performances and allows for system benchmarking. With limited resources, these criteria are particularly important for small and medium-sized enterprises.

Integrating AI with Performance Metrics

Artificial intelligence supports emd-to-end digitization of supply chain. With large datasets from connected assets, AI derives insights and performs tasks more efficiently. The neurile architecture and high volume of data create a natural framework for AI scalability. AIs impact on supply chains exceeds that on many other business areas and is estimated at US$ 2 trillion annually. Much of this value remains untapped because legacy planning and execution systems are overwhelmed by volume, velocity, and variety of data in modern supply chains. The impact is greatest where uncertainty is highest and responsiveness most critical. 

Data-Driven Decision Making

Supply chains operate under conditions of uncertainty that generate substantial risk. Forecasting demand under such conditions is difficult given the stochastic nature of fluctuation. Consequently, inventory optimization has emerged as an effective means of risk mitigation. Traditionally the central concern for supply chains characterized risk, there has been little emphasis on the role of information networks. However, the widespread adoption of digital technologies has created the opportunity to establish effective interorganizational information networks with supply chain partners. To articulate their impact managers must reconsider the metrics with which performance is measured, moving the emphasis beyond cost efficiencies. The ability of artificial intelligence to extract insights from large datasets simultaneously amplifies the value of information networks, transforming the scale and scope of available data. Contemporary research characterizes supply chains as physical, informational, and financial flows, yet existing performance metrics focus predominantly on the former.

Enhanced Forecasting Techniques

Forecasting has historically relied on extensive preprocessing of data, summarizing information, and requiring analysts to have far-reaching knowledge to determine appropriate methodological changes. The literature contains a swath of forecasting techniques based upon statistics, machine learning, and artificial intelligence, applied to various challenges such as sales, production, inventory, or supply. Depending upon the availability of historical data, these techniques may require significant training or recalibration. Because different supply chains exhibit different combinations of maturity and digitization, organizations seek flexible solutions that can concurrently provide traditional guidance and progress to an autonomous, artificial intelligence–driven state.

Today’s multienterprise digital supply chain spans components, equipment suppliers, distributors, manufacturers, packaging companies, logistics providers, and retailers, with each participant managing business processes, demand signals, forecast data, production plans, or inventory allocations. The supply chain represents a complex system of systems in which relevant information may reside, and decision-making does not follow a sequential pattern but is distributed across parties. Demand changes can be incorporated late—commonly called the last-mile problem—but advanced digital platforms can enable demand changes to be propagated upstream in real time (Abdu Alomar, 2022). Large data volumes and commoditized computing power enable sophisticated forecasting models. Data sharing at the supply chain level extends forecasting models from the demand ledger to encompass production, suppliers, operations, and market conditions (Stefanovic, 2014). Segmenting the forecast across various products, customers, and KPMG2017 groups increases computational complexity, but algorithms still allow parameters to be established within minutes, once the data has been collected and refined.

Continuous-term predictive capabilities, which depend on frequently updated data environments, enable organizations to overcome the inertia and latency of monthly or weekly planning. They facilitate autonomous rate adjustments or market-control algorithms and allow detailed simulations of alternative supply and demand scenarios, giving organizations time to respond to new market requirements. Such models can reduce demand forecast errors by more than 30 percent, of capacity requirements by 40 percent, and of transportation and distribution by 39 percent.

Case Studies of Successful Implementation

The retail industry offers a compelling example of successfully implementing new supply chain performance metrics in the digital age. Retailers use metrics such as order cycle time, order fulfillment lead time, and a composite measure known as the Perfect Order to evaluate efficiency and customer satisfaction. Very large retailers often integrate a precise measure of “on-shelf availability” into the Perfect Order metric, which is calculated electronically through analysis of Point-of-Sale data when transactions are stocked out (Zapke, 2019). AI technologies assist in maintaining on-shelf availability by powering automated shelf-scanning robots and handheld shelf audit tools and by enabling sophisticated demand forecasting and dynamic pricing models. These tools grealy improve accuracy of metrics and support real-time decision making that elevates retail performance.

Retail Sector Innovations

Beginning from the ascent of  chain store development in the 1980s, management of supply chain in the retail sector has been chiefly concerned with cost reduction (Zapke, 2019). Much of the effort was directed at developing store shelving capacity and cross-docking facilities to support high inventory turnover – because that was what management was measured on; further investment could also be justified on the grounds that “if demand turns up we will have a product to sell”. AI enabled optimization allows focus to shift from efficiency to agility (Abdu Alomar, 2022).

Manufacturing Sector Transformations

Artificial intelligence transforms the manufacturing sector toward a fully flexible, automated, and cost-efficient supply chain through the digitalization of production processes. Smart manufacturing application areas include dynamic scheduling, quality management, human activity recognition, predictive maintenance, waste reduction, and inventory management.

According to Ebadi, et al. (2021), machine learning and artificial intelligence integration in the manufacturing industry introduces significant challenges to supply chain performance measurement. One major challenge arises from suppliers’ differing natures of AI application, along with the use of cutting-edge technologies across the AI value chain, which affect both the structure and business processes of partner companies. A retail company, for example, must consider a supplier’s implementation of vision systems or automatic robotic arms in assessing its associated costs and overall supply chain performance.

Logistics and Distribution Enhancements

Over the last decade, many companies have relocated production and increased reliance on offshore suppliers. Roughly 80 percent of these products are transported by vessels and container ships, with a significant proportion arriving via ports. These shifts have resulted in increased inventory in transit, longer cycle times, and escalated supply chain risk (Zapke, 2019). The emphasis on cost and service levels has led to oversized fleets, resulting in poor running costs and under-utilization of assets. Navigating through economic downturns, companies seek flexibility and speed to market, but traditional metrics fail to support these objectives. Specific logistics enhancements are necessary to address these challenges.

Enhancements in inventory management include accurate inventory record-keeping, advanced allocation methods, and control and planning of work-in-process (WIP). Accurate inventory records enable the use of virtual inventory across the network, potentially reducing safety stocks significantly and enabling agility in meeting customer demand. Advanced allocation methods prioritize and allocate customer orders based on best-fit objectives and overall routing considerations, particularly effective when machines or equipment are bottlenecks. Control and planning of WIP rely on the principle that average WIP equals average release rate multiplied by flow time; this relationship helps maintain desired WIP levels to meet throughput targets, using distribution-based controllers rather than strict physical inventory counts (Abdu Alomar, 2022).

Improvements in distribution practices involve flow stabilization and process simplification. Flow stabilization avoids unnecessary disruptions, manages workload and capacity, and establishes buffer stocks to disentangle demand-supply interdependencies. Process simplification reduces lead times through fewer activities, reduced setup time, and increased lot sizes by eliminating unnecessary processing steps.

Multi-product scheduling benefits from anticipated demand, social media trends, and weather information to forecast market demand, allowing better preparation and agility in scheduling. These indicators provide insights into potential shifts in customer preferences and consumer characteristics. 

Challenges in Adopting New Metrics

Despite promising experiences and a positive outlook for a broader integration of new performance metrics in supply chains, significant challenges persist. The need to transform existing notions of added value to capture the new digital dimensions of supply chain performance is paramount (Abdu Alomar, 2022). Changing established value paradigms is inherently difficult and has increased in importance with the advent of advanced technologies like AI (Zapke, 2019). Efforts to address these challenges must consider the amount of work involved, the ability to reconfigure supply chains, cultural dimensions, technological constraints, data privacy issues, and employee skills gaps. Transitioning away from traditional metrics requires adopting an approach that focuses on universality and speed and embraces a broader set of metrics. Digital integration offers potential for alignment and consistency but calls for careful evaluation of benefits and costs. The durability of new metrics depends on their consistency and ability to dominate other performance measurements when multiple channels operate in parallel within futures supply chains.

Cultural Resistance

Many organizations acknowledge the importance of measurement to any process. Measurement increasingly identifies what cannot otherwise be determined. Some organizations may be stuck by sticking with the measurements they know, but many recognize that unless there is a way of measuring something it is difficult for it to be promoted and become widespread. When production or quality is low, the cause is often labour (Nozari et al., 2022). When operations fail, it is commonly said that people did not do their job, and yet the wherewithal to measure is often absent. The issues of human skills and enhancement of performance proliferate.

Barriers to the adoption of new paradigms are well known. They exist as a matter of organizational survival. Any surviving organization is unlikely to abandon a paradigm and adopt a new one unless prodded by external forces. When the external world changes, the organization may be forced to change along with it. The dominant paradigm is seldom challenged until the changes are so great that adherence to the old paradigm would ultimately lead to the organization’s extinction. Any organization that advocates a paradigm change will face cultural resistance generated from those devoted to the current paradigm. Pronounce the introduction of robotics or artificial intelligence and some will shout that people will be out of a job. Few say with absolute certainty that the adoption of a new paradigm will improve the organization, and some will express skepticism concerning such a claim.

Data Privacy Concerns

Data privacy presents a significant hurdle to adopting data-driven methods across supply chains. Initiatives to improve predictive analytics frequently call for increased data access and distribution. Yet privacy concerns related to data confidentiality, intellectual property, and general commercial sensitivity continue to deter information sharing among partners (Schoepf et al., 2023). Even when data is disseminated, it tends to be aggregated and anonymized—a practice that often reduces its analytical value and precludes more comprehensive inference. Consequently, organizations must leverage predictive analytics with incomplete visibility into their supply-chain networks and rely on limited data from other tiers.

Acknowledgment of privacy as a barrier has spurred the development of privacy-preserving methodologies. Nevertheless, privacy constraints are, by their nature, most restrictive at the process level—where the most valuable and specific data reside. Several techniques, including data obscuration (such as differential privacy), secure multiparty computation, federated learning, and distributed data spaces, have therefore been proposed to enable analytics while safeguarding sensitive information. These methods, however, depend heavily on direct process-level data access, which supply-chain participants are often unwilling to provide.

Technological Barriers

The rapidly increasing availability of ambient data and the introduction of new decision-making techniques based on artificial intelligence have fuelled the transformation of supply chains (Zapke, 2019). The scale and diversity of operational data provide the basis for a detailed analytical view of the processes and activities. The introduction of Things (IoT) and cloud computing, enables real-time information and new decision-making paradigms (Nozari, et al., 2022). The associated benefits have increased the acceptance of these techniques, which become a competitive advantage. Digitalization is driving a rethink of the criteria to evaluate a Supply Chain partner. The traditional focus on efficiency is now replaced by a wider view that includes flexibility, innovation and sustainability.

Future Trends in Supply Chain Metrics

Innovation and technologies have disrupted supply chains at an unprecedented pace. The new landscape blockchain, the Internet of Things (IoT), machine learning, 3D printing, waste reduction, and cloud computing. This broad spectrum generates an equally complex field of promise, expectation, and reality. Despite current challenges, interest in artificial intelligence (AI) by supply chain organizations continues to grow. AI sits at the center of the Fourth Industrial Revolution unfolding for supply chains worldwide (Zapke, 2019). Organizations should take a proactive approach toward experimentation, pragmatism, and implementation, particularly when addressing the limitations of traditional supply chain performance metrics.

The emergence of AI and other disruptions warrants a fundamental reconsideration of supply chain performance metrics. Traditional metrics, underpinned by cost efficiency, service levels, and inventory turnover, reflect an industrial-era paradigm focused on delivery volumes and deadlines. The digital age demands a broader conceptual approach better aligned with modern operations and strategic priorities. Although the world never quite abandoned the original framework, the extent of its outdatedness has grown. Innovation and transformation amplify limitations: inaccuracy, lack of timeliness, excess complexity, inability to capture end-to-end realities, and most critically, a failure to generate evidence-based insights for decision making. The problem is embedded in two underlying issues. First, the inability of legacy mechanisms to capture the fundamental characteristics of digital-era supply chains. Second, the growing irrelevance of traditional paradigms for the economy, market, and society at large.

Reliable metrics exist, but they often fail to satisfy the requirements of the digital age. Inventory remains a valuable proxy for supply chain health; measure costs; then measure the opposite of costs—customer service or order fulfilment. Yet supply chains in the digital age cannot rely upon universal metrics applicable to all supply chains, everywhere. Instead, performance metrics must align with corporate and operational priorities, structural realities, and specific objectives. Efforts focus in two areas. First, a concentration upon time, velocity, and agility, characteristics undoubtedly central to the digital-age supply chain. Second, the adoption of a customer-centric paradigm that shifts the emphasis from efficiency to responsiveness and satisfaction. This perspective offers a considerably wider space for different formulations and a clearer portal toward digital-era innovation.

Blockchain Integration

Blockchain technology is emerging as game changer for next-level integration of monitoring technologies, digitization, and autonomization within supply chains. The technology facilitates secure and trustworthy inter-organizational relationships through the use of distributed digital ledgers based on cryptography and consensus mechanisms. It can serve as a means for reliable data transmission and sharing in digital supply chains. A certificate of custody, along with a certification process for the recorded data, is fundamental to warrant non-repudiation, trust, and confidence, especially when data that are part of blockchain transactions are stored off-chain.

Blockchain could help supply chains by serving as a certification body that ensures the integrity, authenticity, and provenance of off-chain data. In order to accomplish this, a software connector module has been created that permits communication between an Ethereum-like blockchain and an enterprise information system. Despite the fact that supply chain and blockchain research is still in its early stages, there are uses and advantages that could persuade supply chain managers to embrace the technology and work in a setting where mutual trust exists.

IoT and Supply Chain Visibility

The Internet of Things (IoT) addresses the connectivity between physical products, machines, and transportation, creating a web of information exchange across supply chains and within organizations. This connection allows real-time visibility into inventory, assets, and other logistics information. IoT systems can onboard billions of devices, each with unique identifiers, accessible through standard communication protocols. Plus, they function independently of the Internet. Consequently, IoT enables fully automated, largely self-managed systems (Nozari et al., 2022).

CONCLUSION

Supply chains in the AI and digital era demand a radical shift from traditional performance metrics with a narrow focus on efficiency. Even though cost, service levels and inventory turns are still important topics, the former considerations are not enough to model the level on complexity, dynamism and strategic priorities of modern supply chains. Emerging metrics, focused on agility, customercentricity, and sustainability, provide a more holistic, long-view approach that bridges operational realties with the global need for resilience, innovation, and accountability. AI and digital technologies are what make these new measures not just possible, but necessary, because of the predictive capabilities, the automation, the real-time decision-making at a scale never before possible. Finally, a re-examination of SC performance metrics is not an academic exercise; it is a business necessity for organizations wanting to succeed in a turbulent, digitally intermediate world.

REFERENCES

1. Acquaye, A., Ibn-Mohammed, T., Genovese, A., Afrifa, G. A., Yamoah, F., & Oppon, E. (2017). A quantitative model for environmentally sustainable supply chain performance measurement.
2. Alomar, M. A. (2022). Performance optimization of industrial supply chain using artificial intelligence. Computational Intelligence and Neuroscience, 2022(1), 9306265
3. Archana, T., & Stephen, R. K. (2024). The Future of Artificial Intelligence in Manufacturing Industries. In Industry Applications of Thrust Manufacturing: Convergence with Real-Time Data and AI (pp. 98-117). IGI Global Scientific Publishing.
4. Arredondo, C. R., & Alfaro Tanco, J. A. (2021). Supply chain management: Some reflections to improve its influence in business strategy. Innovar, 31(81), 7-19.
5. Chatterjee, S., & Chaudhuri, R. (2022). Supply chain sustainability during turbulent environment: Examining the role of firm capabilities and government regulation. Operations Management Research, 15(3), 1081-1095.
6. Czinkota, M. R., Kotabe, M., Vrontis, D., & Shams, S. R. (2021). Distribution and supply chain management. In Marketing management: Past, present and future (pp. 499-552). Cham: Springer International Publishing.
7. Deng, Q., Huang, X., Zou, J., & He, Y. (2024). Screening of sustainable supply chain performance evaluation indicators based on the ill-conditioned index cycle method. Plos one, 19(3), e0293038.
8. Ebadi, A., Xi, P., Tremblay, S., Spencer, B., Pall, R., & Wong, A. (2021). Understanding the temporal evolution of COVID-19 research through machine learning and natural language processing. Scientometrics, 126(1), 725-739.
9. Ghodake, S. P., Malkar, V. R., Santosh, K., Jabasheela, L., Abdufattokhov, S., & Gopi, A. (2024). Enhancing Supply Chain Management Efficiency: A Data-Driven Approach using Predictive Analytics and Machine Learning Algorithms. International Journal of Advanced Computer Science & Applications, 15(4).
10. Gunasekaran, A., Yusuf, Y. Y., Adeleye, E. O., Papadopoulos, T., Kovvuri, D., & Geyi, D. A. G. (2019). Agile manufacturing: an evolutionary review of practices. International Journal of Production Research, 57(15-16), 5154-5174.
11. Herlitzius, T., Noack, P., Späth, J., Barth, R., Wolfert, S., Bernardi, A., … & Jakob, K. (2022). Technology perspective. In Handbook Digital Farming: Digital Transformation for Sustainable Agriculture (pp. 109-189). Berlin, Heidelberg: Springer Berlin Heidelberg.
12. Koundal, P., & Bhalla, M. (2025). The Role of AIS in Enhancing Small Business Agility and Sustainability in Industry 5.0. In The Future of Small Business in Industry 5.0 (pp. 301-330). IGI Global Scientific Publishing.
13. Licardo, J. T., Domjan, M., & Orehovački, T. (2024). Intelligent robotics—A systematic review of emerging technologies and trends. Electronics, 13(3), 542.
14. Mohamed, A. E. (2024). Inventory management. In Operations Management-Recent Advances and New Perspectives. IntechOpen.
15. Mubarik, M. S., & Khan, S. A. (2024). Future of digital supply chain management. In The Theory, Methods and Application of Managing Digital Supply Chains (pp. 163-178). Emerald Publishing Limited.
16. Nozari, H., Szmelter-Jarosz, A., & Ghahremani-Nahr, J. (2022). Analysis of the challenges of artificial intelligence of things (AIoT) for the smart supply chain (case study: FMCG industries). Sensors, 22(8), 2931.
17. Olaniyi, O. A., & Pugal, P. S. (2024). Optimising inventory management strategies for cost reduction in supply chains: A systematic review. Jurnal Akuntansi dan Bisnis: Jurnal Program Studi Akuntansi, 10(1), 48-55.
18. Perano, M., Cammarano, A., Varriale, V., Del Regno, C., Michelino, F., & Caputo, M. (2023). Embracing supply chain digitalization and unphysicalization to enhance supply chain performance: a conceptual framework. International Journal of Physical Distribution & Logistics Management, 53(5/6), 628-659.
19. Pourmajidi, W., Zhang, L., Miranskyy, A., Steinbacher, J., Godwin, D., & Erwin, T. (2021). The challenging landscape of cloud monitoring. Knowledge Management in the Development of Data-Intensive Systems, 157-189.
20. Qian, X., & Papadonikolaki, E. (2021). Shifting trust in construction supply chains through blockchain technology. Engineering, Construction and Architectural Management, 28(2), 584-602.
21. Rana, J., & Daultani, Y. (2022). Mapping the Role and Impact of Artificial Intelligence and Machine Learning Applications in Supply Chain Digital Transformation: A Bibliometric Analysis. Operations Management Research. https://doi.org/10.1007/s12063-022-00335-y
22. Schoepf, S., Foster, J., & Brintrup, A. (2023). Identifying contributors to supply chain outcomes in a multi-echelon setting: a decentralised approach.
23. Stefanovic, N. (2014). Proactive supply chain performance management with predictive analytics. The Scientific World Journal, 2014(1), 528917. Stefanovic, N. (2014). Proactive supply chain performance management with predictive analytics. The Scientific World Journal, 2014(1), 528917.
24. Tuni, A., & Rentizelas, A. (2019). An innovative eco-intensity based method for assessing extended supply chain environmental sustainability. International Journal of Production Economics, 217, 126-142.
25. Yerra, S. (2025). Optimizing supply chain efficiency using AI-driven predictive analytics in logistics. International Journal of Scientific Research in Computer Science Engineering and Information Technology, 11(2), 1212-1220.
26. Zapke, M. (2019). Artificial Intelligence in Supply Chains (Master’s thesis, Universidade NOVA deLisboa (Portugal)).