Management of supply shortages with AI in crisis

Research Project with focus on resilience and risk management of supply chains especially in the health and rescue sector using a data plattform

Logo Research Projekt ResKriVer
© Fraunhofer IML

Research Proposal

ResKriVer aims to

  1. Identify and counteract early in situations of supply shortages due to crisis and
  2. Build a transparent communication for society in case of a crisis.

Especially the corona outbreak made us aware of consequences caused by crisis. Worldwide, we faced delivery shortages, irrational panic shopping and interruptions of transport routes. This has impacts on local as well as global supply networks.

Problematic situations have to be accordingly detected, evaluated and abstracted in order to be able to respond to crises, emerging disasters or imminent risks. Only this way, a unified process mechanism can be used on the upcoming problem and initiate an appropriate response to it. To identify and evaluate risks, their impact on suppliers networks and helpful and effective measures, there is a need for innovative methods, that are investigated and collected within ResKriVer.

 

Market perspective and product promises

An important issue of ResKriVer consists in prevention and precaution of crisis to avoid consequential costs and damage. An important aspect besides the maintenance of suppliers structures (energy, gas, water) and communication networks are measures, which ensure the supply of crisis-relevant goods such as medical products, medicines and technical equipment in the event of a pandemic.

The basis of ResKriVer is a digital platform that enables the collection, formation and communication of crisis-relevant information about supply chains with help of AI applications. With collections and evaluations of crisis-relevant data using AI methods, this platform should enable bottlenecks in the supply chains of companies and public demanders to be sufficiently predicted in advance so that appropriate measures can be initiated. In the next step, crisis information is transmitted efficiently and appropriate to the target group. For this purpose, the ResKriVer platform provides a range of crisis communication tools.

The Aim is a broad use of the data platform for the health and emergency services as well as for companies - especially for SMEs - in order to enrich their risk management in crisis situations with near-real-time data and to show necessary alternative courses of action.

Overview of content correlations within ResKriVer

data, platform and usage of ResKriVer node
© Fraunhofer IML

The role of Fraunhofer IML

Fraunhofer IML contributes risk management expertise and the supply chain simulation tool OTD-NET to ResKriVer. By coupling the simulation tool with AI approaches, Fraunhofer IML is assessing crisis situations and generates suggestions to guarantee supply chain resilience (Viable Supply Chain). This evaluation of crisis situations is divided into three steps. The first step, the "AI Status Rating", is about the determination of whether a crisis event has an effect on the supply chain. If bottlenecks exist, relevant parameters are determined and adjusted in the second step, the "AI Scenario Creator". This results in various scenarios that are intended to compensate for the effects of a crisis event. By means of a feedback loop, the scenarios are simulated in the "AI Solution Assessment" and fed back to the “AI Scenario Creator”. The goal of this assessment is to evaluate different scenarios and select the best option.

Example of a crisis’s global outreach

example of risk impact on supply chain, ResKriVer
© Fraunhofer IML

Challenge and innovation

One of the goals of ResKriVer is to expand digitization in the economy and thus increase Germany's economic strength. To implement the planned ResKriVer platform with all the necessary data, a high degree of digitization is required, which has only existed to a limited extent so far. One challenge will therefore be the creation of a digital twin that remains constantly expandable and adaptable in order to be able to predict the effects of different events on the supply chain.

 

In keeping with the project's mission, special consideration will be given to the security and confidentiality of sensitive data. The AI solutions as well as the digital platform’s specific design meet specific security criteria to enable data exchange under secured conditions. Furthermore, the findings of the crisis-specific models will be visible to the population and crisis teams via the platform. By means of a suitable form of presentation and communication, recommendations for action are to be communicated here and the foundation stone for a higher acceptance on the part of the population and on the part of the organizations is to be ensured.

 

Partners

Fraunhofer-Institut für Offene Kommunikationssysteme (FOKUS), Berliner Feuerwehr, Charité - Universitätsmedizin Berlin, Condat AG, eccenca GmbH, Fraunhofer-Institut für Materialfluss und Logistik (IML), HFC Human-Factors-Consult GmbH, KomRe AG, Merantix Labs GmbH, Rundfunk Berlin-Brandenburg (rbb), Vereinigung zur Förderung des Deutschen Brandschutzes (vfdb) e.V., YOUSE GmbH