Planning delivery routes and disposition with AI

Workforce and asset management from Windeck 

GreenGate AG, based in Windeck in North Rhine-Westphalia, is a software provider founded in 2020 specialising in operations management and maintenance solutions (workforce and asset management).  The core areas of activity are the development and distribution of software and hardware (including associated services) and consulting, as well as conducting training courses and seminars. Many of their customers are regional utility companies, such as energy grid operators, who use electric vehicles, for example when technicians replace a household's electricity meter. The aim is to make these deployments as effective and environmentally friendly as possible by considering relevant processes and customer needs. 

Appropriate maintenance of electric vehicles 

Current studies indicate that by 2030, there will be a significant shift in the types of automotive drive systems - from the combustion engine to alternative drive systems. Electric vehicles offer regional service providers an opportunity to fulfil their service areas in a more cost-effective and environmentally friendly way. However, the service lifespan of electric vehicles is heavily dependent on the appropriate maintenance of the battery and is therefore sensitive to route planning, usage frequency and intensity, and driving behaviour. 

Mission control tool for electric vehicles 

An AI-based decision support system is being designed and prototypically implemented to facilitate and improve the use of electric vehicles. To this end, the pilot project of the Green-AI Hub Mittelstand is investigating the use of advanced AI to improve usage planning and route optimisation. This is done by overcoming the limitations of conventional single-point predictions through uncertainty quantification (UQ), e.g. by including certain uncertainty factors into the predictions. This integration is intended to increase the transparency and reliability of the models and reveal correlations between route planning, driving behaviour, and environmentally conscious vehicle use. Methods of uncertainty quantification are being researched, including ad-hoc techniques and conformal predictions. In addition, methods are used to explain how AI arrives at its predictions, so-called ‘eXplainable AI’. This helps the decision-makers and planners better understand how the AI works with model predictions, avoiding issues like defective batteries or delays. 

Optimised process and route planning 

By using advanced machine learning and explainable AI methods, GreenGate AG's customers will be able to perform smarter and more efficient scheduling and route planning in the future. The uncertainty analysis integrated into the ML modelling process provides deep insights into model reliability, thereby increasing decision transparency. This increases the efficiency and effectiveness of the entire process chain. In addition, the methods developed help to avoid problems such as premature battery defects or delays in order fulfilment. This would be a significant added value for GreenGate AG's customers.