blog Article

Joris Roels Joris is a Machine Learning Engineer and Solution Architect at Radix. He obtained a Master's degree in Mathematical Computer Science and a Ph.D. in Engineering at Ghent University. After eight years of research in Computer Vision and Artificial Intelligence he brings a keen interest in math innovation and socially relevant problems onboard. Joris also worked as a visiting scientist at Heidelberg University for five months.

Revolutionizing Value Chain Management with AI

Author: Maria Merenda ,

Today’s world is highly unpredictable. Global events (COVID-19 pandemic, the blockage of the Suez channel and the geopolitical situation, etc. ) as well as local scale events such as variable customer demand, supplier delay and equipment failure make it increasingly difficult for companies to manage their value chain efficiently. 

In this article, we present examples of value chain challenges faced by various industries, and explain how AI tools, such as our newly developed software framework, address them. 

Let’s start with the obvious: what does “value chain management” actually mean? Value chain management (VCM) refers to the process of coordinating and optimizing all activities involved in designing, producing, delivering, and supporting a product or service. 

In real-life applications, VCM faces several challenges. These challenges include:

  • Supply Chain Complexity: Modern supply chains are often complex and global, involving multiple suppliers, manufacturers, distributors, and retailers. 
  • Demand Volatility: Fluctuations in customer demand can create uncertainty and disrupt the entire value chain if not properly anticipated.
  • Inventory Optimization: Balancing inventory levels to meet customer demand while minimizing holding costs is a critical challenge. 
  • Operational Efficiency: Optimizing production processes, reducing waste, improving logistics and transportation are key factors in achieving operational efficiency.
  • Sustainability and Compliance: Increasingly, businesses are under pressure to adopt sustainable practices and comply with regulations related to environmental, social, and governance (ESG) standards.

How can AI address these challenges?

Artificial Intelligence (AI) has emerged as a powerful tool to address the complex challenges faced in value chain management. With its ability to analyze vast amounts of data, AI enables organizations to make accurate forecasts, develop robust planning strategies, monitor operations in real time, mitigate risks, and optimize post-operation activities.

Radix framework

To make this a little bit more tangible, the next section will illustrate the challenges in operations management faced by various industries and how, by linking different AI applications together, this process can become more automated, insightful, and cost-efficient.

Real-world examples

Each industry has its unique set of challenges and requirements. In manufacturing, VCM focuses on streamlining production processes, managing suppliers, and ensuring timely delivery of goods. Retail and distribution industries leverage VCM to optimize inventory levels, improve order fulfillment, and enhance logistics operations. Additionally, industries such as healthcare, agriculture, technology, and finance also employ VCM to optimize their operational workflow and enhance overall performance.


A supermarket offers a wide variety of perishable products, such as fruits, vegetables, meat, fish, pre-made meals, and more. 


Every day the store manager needs to deal with products that are close to being perished and remain unsold, while some other products are out of stock, resulting in waste, loss of revenue, and customer dissatisfaction. 


Let's explore how different AI solutions can support store managers in addressing various challenges. Firstly, an AI forecasting model can analyze purchase history, store traffic, holidays, and promotional campaigns to anticipate customer behavior, enabling store managers to order the right products at the appropriate time. Secondly, AI can automate employee scheduling, replacing ad-hoc decisions and allowing for smoother transitions between tasks, such as employees switching from filling shelves to operating the counter based on store needs. Thirdly, AI technologies can monitor operations by analyzing input from cameras and other devices, detecting anomalies like empty shelves or long queues. Additionally, an AI copilot can mitigate certain product shortages by, for example, updating next-day orders to wholesalers as needed.

In an ideal scenario, all these AI solutions would be interconnected, influencing each other. Imagine a world where empty shelves trigger new employee schedules to ensure timely restocking or where orders automatically adapt to current stock levels. Furthermore, employee breaks could be planned based on weather forecasts, assuming that customers are more likely to visit the shop after the rain has stopped. The combination of customer behavior forecasting and real-time monitoring empowers supermarkets to place accurate orders with suppliers and optimize staff planning, resulting in efficient operations.


In this example, we will consider a printing company whose core business is to perform a large number of interdependent printing jobs (the actual printing, cutting, assembling, …) on various machines, for many different customers.


The printing company performs a lot of custom printing jobs. Resulting in a very complex non-standard printing schedule. Today these jobs are planned manually, resulting in non-optimal and often even infeasible schedules. Next to this, last-minute urgent orders from customers and variable suppliers' lead times make it difficult to schedule the different printing tasks in a cost-efficient way while delivering orders on time.


Let's again explore how different AI solutions can assist the planning manager in addressing challenges. AI-driven forecasting models utilize current order pipelines and historical data to provide insights into future customer orders and supplier lead time. With AI-based planning, cost-efficient production plans can be created, considering order intake, available resources, and production constraints, optimizing task assignments to minimize time. Real-time monitoring, supported by predictive maintenance algorithms and tracking systems, enables the identification of anomalies and allows for timely interventions. In response to real-time events, like machine breakdown, the AI copilot can mitigate by generating production plans that meet new requirements while deviating minimally from the original plan.

Ideally, these elements work together in a virtuous circle, where accurate forecasting benefits planning, monitoring enables mitigations, and the quality of future forecasts improves, enabling continuous learning and better decision-making.


This use case focuses on a road parcel delivery company that serves a high volume of customers every day all over Europe.


Planning the routes of the trucks in a cost-efficient way while delivering the parcels in time is challenging, even on paper. In reality, the parcel company also has to deal with variable demand, unexpected traffic jams, and truck breakdowns.


In exploring the potential of diverse AI solutions, let's uncover how they synergistically assist the planning manager in effectively tackling the current challenge. Through accurate parcel quantity forecasting and traffic prediction, AI models enable informed planning decisions. Utilizing advanced operation research algorithms, the planner assigns parcels to trucks, optimally plans routes, and factors in operational parameters like fuel consumption and maintenance costs. With comprehensive monitoring systems analyzing GPS trackers, sensors, and RFID tags, anomalies such as delays, route deviations, or temperature variations can be swiftly identified and addressed to mitigate disruptions. The AI copilot adapts by automatically generating updated routing solutions, ensuring timely pickups and deliveries. 

The seamless integration of forecasting, planning, monitoring, and mitigation empowers the logistics company to holistically optimize operations and ensure prompt parcel shipments.

How Radix can help

Throughout our journey of assisting prominent companies in solving intricate value chain situations, we have been driving and experiencing firsthand the transformative shift that comes with integrating AI solutions. Some of our clients have already achieved significant improvements in on-time deliveries, resource allocation, planning efficiency, and customer satisfaction. Leveraging the knowledge gained from these collaborations, we have developed a software framework designed to aid our clients in building their own digital twin, effectively co-piloting their value chain.

Our framework offers a combination of easy-to-use, customizable, and combinable components that harness the synergies of machine learning and operations research. These components encompass techniques such as time series analysis to gain insights into demand variations and enhance planning workflows, as well as state-of-the-art root cause analysis algorithms to identify operational bottlenecks. As AI continues to revolutionize value chain management, companies that embrace this technology will gain a competitive edge, adapt to dynamic market conditions, and unlock new levels of success in today's fast-paced business landscape.

Interested to learn more or ready to start brainstorming on what this could mean for your organization? Feel free to contact us, and we will be happy to help!

Maria Merenda
About The Author

Maria Merenda

Maria's goal, as Solution Lead at Radix, is to build a long-lasting relationship with our clients and guide them to reach business transformation by using Al technology. With a Master's degree in Business Engineering and 9 years of work experience at Siemens, Maria is one of our latest additions to the team.

About The Author