Harnessing value in unstructured data to surpass traditional AI’s limitations
Generative AI (GenAI) has quickly become the technology to watch — one with the potential to overhaul industries. In supply chain management, it holds particular promise for making sense of large volumes of unstructured data. Recent global trade disruptions, such as the sudden US tariff changes, highlight GenAI’s relevance and the limits of traditional AI in fast-moving environments. This Viewpoint evaluates whether GenAI meets the hype using a standardized supply chain model and identifies where it can add the most value.
Global supply chains are under pressure, facing rising complexity, less predictable risks, changing demands, and the relentless need for efficiency. This has prompted companies to seek innovative optimization solutions. Traditional AI has been a game changer through rule-based systems, machine learning (ML), and automation. Widespread adoption, however, remains uneven, with many companies still in the early stages of implementation. And its limitations are increasingly apparent: traditional AI depends on stable, structured data and struggles with fast-moving, unforeseen developments. GenAI is now broadening its application. Unlike its predecessors, it not only analyzes data but can redefine decision-making and content creation across the supply chain. Can GenAI truly transform the way supply chains operate?
To assess GenAI’s potential impact, we begin by distinguishing between traditional AI and GenAI and follow with an evaluation of its potential for supply chains:
The Supply Chain Operations Reference (SCOR) model is a standardized framework used to evaluate, benchmark, and improve supply chain performance (see Figure 1). It provides a comprehensive structure for analyzing supply chain processes across six main processes — plan, order, source, produce, fulfill, and return — which, together orchestrate end-to-end integration.
In principle, GenAI can add value across multiple points in the supply chain. The key question, however, is where and how it delivers the greatest impact. Arthur D. Little (ADL) has identified those processes where the benefit of GenAI is greatest, along with advisable implementation.
In the plan phase of the SCOR model, demand forecasting is vital to ensure that supply chain operations align with market needs. Traditional AI has long supported this task by analyzing historical sales data to identify patterns and trends. However, in volatile and rapidly changing markets — where demand is influenced by external factors such as geopolitical events, economic disruptions, or emerging consumer trends — traditional AI’s reliance on past data often falls short. This limitation was evident during recent global disruptions, such as US-China trade tensions around tariffs, and unprecedented events like pandemics and extreme weather, which are difficult to model using traditional structured data alone.
GenAI addresses these gaps by synthesizing real-time, unstructured data from sources such as social media, breaking news, and consumer behavior. The result is sharper, more adaptive forecasts when historical patterns no longer apply. For example, during the COVID-19 pandemic, traditional AI models struggled to predict demand accurately because past sales patterns were no longer relevant. Here is where GenAI and traditional AI can complement each other: GenAI can preprocess data, such as summarizing news and trends, before traditional AI performs in-depth analysis and forecasting; GenAI can then generate user-friendly, insightful outputs.
A cutting-edge enhancement is persona-driven forecasting through agentic AI. By simulating personas representing diverse customer segments, AI agents incorporate nuanced behavioral insights. Combining all personas provides a comprehensive view of market needs (e.g., based on current trends in news or social media sentiments), allowing more granular customer segmentation and forecasts that better reflect heterogeneous market needs and consumer preferences.
In ADL’s experience, almost half of its clients have integrated ML into their demand forecasting. In consumer goods, GenAI improves forecast accuracy by examining historical data alongside external factors such as weather and market trends. Across multiple clients, GenAI has shown potential to increase accuracy by 15%-30%, translating into a 5%-10% reduction in inventory costs by extending traditional data processing to include sources like social media sentiment or image data. These figures exclude administrative efficiencies, where GenAI uses unstructured data and enables natural language interaction, making insights more accessible to employees and reducing reliance on specialists.
Case study — Enhancing ice cream–demand forecasting
A leading consumer goods company uses GenAI to enhance ice cream demand forecasting by integrating real-time weather data and image recognition. In one region, forecast accuracy improved by 10%. AI-enabled freezers with visual inventory tracked reduced stockouts, which increased sales by 8%-30% in pilot markets, optimizing inventory and reducing waste, especially for highly seasonal products.
GenAI can be particularly valuable in the order phase, especially for contract management — a complex process involving legal wording and large volumes of documents. GenAI helps procurement teams efficiently understand and manage contracts. It supports automated compliance checks by screening and summarizing clauses to identify risks, critical terms, and problematic items requiring further attention. Additionally, GenAI enhances document tracking and provides automated notifications, giving teams an up-to-date view of supplier and product/service coverage, contract renewals, and upcoming deadlines. This boosts efficiency and allows procurement to fully leverage existing contracts.
Today, over 70% of companies struggle to find and leverage at least 10% of their contracts, according to the Journal of Contract Management. As a result, they often rely on market prices rather than negotiated terms, leading to unnecessary costs. Effective contract management can improve profitability by 9%, as reported by the International Association for Contract & Commercial Management (IACCM). GenAI addresses this gap by ensuring contracts are properly identified and applied, reducing administrative burdens, lowering operational costs, and enhancing supplier relationship management and sourcing efficiency through greater transparency and understanding.
In manufacturing, maintenance-related downtime directly impacts productivity and profitability. In the produce phase, GenAI optimizes predictive maintenance by analyzing extensive records and logs, helping organizations improve operational efficiency. It transforms maintenance data into concise, actionable instructions that technicians and engineers can readily understand and implement. This accelerates repair processes, reduces reliance on skilled workers, and improves first-time fix rates.
It can also generate instructions where none existed and enhance the quality of existing ones, minimizing operational disruptions. By analyzing patterns in maintenance logs and historical equipment data, GenAI allows knowledge sharing across machines and systems, supported by interactive guidance between AI, users, and maintenance experts. Training costs can also be reduced, as GenAI provides on-demand support, partially replacing complex onboarding and training programs.
Case study — Streamlining maintenance for a transmission system operator
ADL developed a GenAI solution for a leading European transmission system operator that autonomously creates maintenance instructions based on ERP (enterprise resource planning) data and maintenance experience. The system also generates and triggers orders directly in SAP. This solution reduced effort in creating instructions by almost 40%, which were previously carried out by skilled engineers.
Effective communication is the backbone of order and return management. GenAI-powered chatbots are revolutionizing interactions with customers and suppliers by providing instant, accurate responses. For customers, chatbots handle common inquiries, provide instant solutions, and personalize answers rather than offering generic responses. Complex inquiries can be escalated to human agents, with AI providing relevant context about the customer. For suppliers, automated responses streamline communication, instantly providing information on invoices, compliance, or inventory. GenAI combines personalized communication with accurate data and collects interaction data to improve supplier management, identifying recurring issues for proactive resolution.
By enhancing customer engagement and optimizing supplier contact, GenAI chatbots provide real-time support, improve customer experiences, and strengthen supplier relationships. Companies can achieve 20%-40% cost savings in customer management by integrating chatbots. Savings vary depending on task complexity, interaction volume, system integration, and the need for human intervention.
Case study — Enhancing customer service & supplier negotiations in retail
A major retail company uses GenAI chatbots to enhance both customer service and supplier negotiations. On the customer side, AI-powered assistants provide real-time support for product inquiries and order tracking, improving user experience and reducing reliance on human agents. On the supplier side, a GenAI negotiation bot automates routine vendor deals, reducing negotiation times from weeks to days and improving terms in over 60% of cases. Chatbots cut support and procurement costs by 20%-30%.
GenAI can be a game changer for supply chain management, but its impact depends on careful application aligned with strategic business needs. Its adaptability and ability to handle complex, unstructured data make it ideal for high-value areas such as contract management, maintenance, and real-time customer interactions while also enhancing existing applications like demand forecasting. It optimizes processes and information handling and frees internal resources to focus on more strategic initiatives, including market analysis, product development, and strengthening supplier relationships.
However, GenAI is not a silver bullet for all supply chain management improvements. In data-driven areas like demand planning or maintenance, for example, it is best to use it in combination with traditional AI. GenAI is well-suited for preprocessing and aggregating data from diverse sources (e.g., social media, news) and generating user-friendly outputs, while traditional AI handles intermediate processing, reducing the risk of biases or “hallucinations” from pattern-seeking algorithms.
GenAI does not transform the entire supply chain at once, but it can significantly optimize specific applications. Overall, the hype surrounding GenAI in supply chain management is warranted — provided it is applied to the right use cases. Companies should assess their needs, context, and risks to identify where GenAI provides a strategic advantage. Success will come from thoughtfully combining GenAI’s strengths with human expertise and traditional AI capabilities.
GenAI is not a universal fix, but when applied with intent, it can transform specific, high-value areas of supply chain management. Its strength lies in handling unstructured data, enhancing decision-making, and unlocking new efficiencies in functions such as contract management, maintenance, and customer interactions. To realize its full impact:
Used strategically, GenAI becomes more than a tool; it’s a multiplier forspeed, insight, and adaptability in next-generation supply chains.
By Dr. Kai-Oliver Zander, Dr. Thomas Thiele, Maurice-André Ruberg, Julia Pilotto, Tim Ahlers
cloud technology axon
AIVORIQ has grown significantly since Alantra PE launched the project....
