< Project Overview >
Autonomous supply chains are an emerging trend with potential to change supply chain management (SCM) practice by eliminating repetitive processes. The Internet of Things (IoT) can play a crucial role in bringing this concept to reality. Currently there are very few industrial adopters of autonomous supply chains despite their promise for reducing unnecessary costs.
This project will build on and integrate a number of current and previous research efforts funded by Research Councils and industrial partners, in the area of autonomous supply chain management systems. It will assess the technical feasibility of deployment in real life scenarios in terms of scalability, trustworthiness, persistence and emergent outcomes of such systems.
Supply chains consist of multiple interacting companies each pursuing their own goal with limited knowledge of the overall picture. This creates a complex system with unique learning outcomes and numerous common challenges, such as finding trade-offs between demand and the costs of satisfying demand under uncertainty.
In the past, the Distribution Information and Automation Laboratory (DIAL) research group at the University of Cambridge’s Institute for Manufacturing (IfM) has worked on creating autonomous decision making and SCM systems that can help individual companies mitigate such problems using a combination of: IoT based data capture, autonomous systems and manufacturing analytics. While these projects developed multiple innovative and successful approaches to autonomous SCM they remain in the academic realm and have not been implemented in industrial settings. Reasons for this include:
- Lack of an integrative approach that brings these siloed projects together to make a comprehensive system
- analysis of how such systems could integrate with company supply chain IT systems
- demonstration of system feasibility, scalability and persistence
This project will address these challenges by integrating research outputs, bringing them to higher technology readiness levels so that transition to industrial settings is possible. Because of the integrative nature of the project, the project is sub-divided into three mini-projects:
1) Autonomous supply chain with intelligent products
Intelligent products are self-organising systems that can sense their current state and plan ahead by communicating their needs through the IoT. Based on previous approaches proposed we will create a self-organising supply chain system demonstrator in which products place orders, communicate with each other to schedule delivery, and monitor delivery performance.
2) Uncertainty prediction with IoT data
IoT based systems can be used to assess the delivery performance of suppliers as data on deliveries will be higher in volume and more granular with each product / object connected to the Internet. We expect that this newly available data can improve predictions on when deliveries will be made, so that planning and scheduling of processes that use supplier products can be improved. This mini-project will build on the previous one to incorporate uncertainty prediction to the self-organising intelligent product system.
3) IoT in provenance
Accurate description of the provenance of products is a big issue that IoT systems may help address. We will demonstrate the use of IoT in achieving better provenance by allowing new products to join in and old products to drop out to the self-organising system developed in mini-project 1. Over time, intelligent products will be generating enough data to communicate with one another to assemble a history of provenance and share that information.
At the end of the development period an industrial workshop will showcase an integrated demonstrator and ask participants to provide the project with further feedback on industrial adoption and how potential barriers can be addressed.
Dr Alexandra Brintrup, Lecturer in Digital Manufacturing at the University of Cambridge.
Industrial partners will be contacted throughout the project to contribute towards workshops.