A demonstrator and reference framework IoT-based supply chain digital twin

24/06/20

< Project overview >

Supply chains emerge as complex networks of interdependent companies. While companies know who their immediate neighbours (buyers and suppliers) are, they often lack full supply chain visibility. Despite this, they must coordinate to deliver goods and services on time, whilst also reducing the overall costs.

This information asymmetry leads to emergent effects such as demand amplification and systemic disruptions, and therefore to wasted materials and time. Methods such as Collaborative Planning, Forecasting and Replenishment require companies to integrate individual enterprise resource planning (ERP) systems, or set up data patches, which require investment, time, and discipline.

Cost becomes a barrier to the exchange of information, and investment becomes a risk, as it perpetually ties companies together, preventing the formation of new alliances. A more fluid, flexible, and decentralised IT solution is needed to overcome information asymmetry.

In this project, our aim was to develop a reference framework for the IoT-enabled improvement on information asymmetries at the supply network level.

The reference framework was be based on the development of a “supply chain digital twin”, which included a data-driven analytics and simulation engine that replicated a real-life supply chain, by using IoT data from goods that were flowing in the chain.

The digital twin was a platform that collected and aggregated real-time data from supply chain partnering organisations, mitigating the need for integrating individual ERP systems. By doing so, network-level data could be used to perform both optimisation of coordination for the current state, and what-if analyses for the future state.

The project objectives

  1. Identify how IoT-based data capture can be used alongside supply chain information systems.

  2. Develop a supply chain digital twin to:

    • Identify behavioural patterns, by applying machine learning methods to multiple data streams.

    • Build realistic agent-based models (ABM) of supply chain networks, and calibrate and validate these models.

    • Enable increased prediction accuracy and optimised coordination amongst supply chain partners.

  3. Evaluate and refine the supply chain digital twin, and the IoT-enabled improvement, through use cases.

  4. Validate and verify results through a demonstrator.

Project leads

Project partners

< Theme >