"By introducing automated route provisioning, we have been able to relieve our drivers of a considerable amount of work."

In a major city like Berlin, efficient waste management plays a crucial role. A key player in this area is Berlin Recycling GmbH, which is using artificial intelligence (AI) to break new ground in waste disposal and recycling. Their AI-supported route planning not only helps optimize operations, but also reduces CO2 emissions. In this interview with Bianka Rieder, CEO of Berlin Recycling, we explore the challenges, opportunities and future of AI in waste management.

Berlin Recycling is dedicated to optimizing waste disposal and recycling processes in a major city like Berlin. Could you give us an overview of how Artificial Intelligence (AI) is used in your processes?
About two years ago, we started thinking about how it would be possible in the future to keep our drivers fit and employed in the company in the long term. On the one hand, it is important to know the exact data of a daily route (e.g., number and size of containers) and, on the other hand, to know the needs of the employees and incorporate them into the planning. With these and other collected data, we were able to "feed" the AI so that it then determines and schedules the best possible tour staffing for the planned tour. However, this is only one part of the story. The overall vision is to use AI to, among other things, predict how "heavy" or "light" a (new) customer will be for our drivers. Thus, we get a kind of classification by load level and can use this classification to build up the tours in a fair(er) way. Ultimately, we want to create a dispatch tour planning system with the help of AI, where the AI ultimately automatically assembles the tour and tour crew that shows the best possible match, from dispatching to operational tour planning.

In a complex urban environment like Berlin, efficient route planning for garbage trucks is certainly a major challenge. How has the introduction of AI changed these aspects of your work and what specific improvements have you already seen?
For example, by introducing automated route provisioning, we have been able to relieve our drivers of a considerable amount of work, as they no longer have to write down their routes by hand on a so-called routing slip, but can simply follow the planned route. This saves time and paper!

Reducing CO2 emissions is a key concern for many cities and businesses. How has the implementation of AI at Berlin Recycling contributed to achieving these goals and how does this integrate with your long-term sustainable development strategy?
The continuous optimization of our tour areas is part of the quality requirements within our processes and was also strictly adhered to before the use of AI. Reducing mileage on tours has been part of our environmental management system for more than 10 years and is continued annually. The use of AI in this area is therefore more of an automation of processes. Of course, it pays off for our sustainability management if we manage to keep our employees in the company for the long term and, in particular, keep them fit.

Every technology introduction brings its own challenges. What hurdles did you have to overcome when implementing AI in Berlin Recycling's workflows and how did this affect your employees?
The skepticism on the part of our drivers as to whether the scheduling of tours and the provision of routes would really work was naturally high at the beginning, but it was reduced through prior talks and discussions, especially with the works council. Particularly in the area of route provision, it was possible to see that the employees were more satisfied, since the annoying "paperwork" of the routing slips was eliminated and they could concentrate better on their work.

In addition, it had to be ensured that the functions would be systemically connected to the ERP system, which proved to be the biggest hurdle so far (time, costs). Therefore, at the beginning we used a BI tool (business intelligence - QlikSense), which the dispatchers can still use today.

The future of waste management and recycling will undoubtedly be heavily influenced by the advancement of AI. What is your vision for the future application of AI in these areas and what specific steps are you taking to realize that vision?
One of our visions is called ITP. ITP stands for intelligent route planning and is intended to contribute to keeping our drivers healthy in the long term during their working lives. What we need for this is more data. Data, for example, on the fitness level of the employees, data on the job site (steps, basements, backyards). Also worth mentioning is the use of AI to group and classify waste to analyze, for example, misthrows or proportion of certain waste fractions. The first tests have already been started for this. The use of AI for image recognition, for example to analyze street damage, defective or graffitied street signs, is also being planned.

The capital has become an important center for innovation and technology in Europe. How do you see the role of Berlin and especially Berlin Recycling GmbH in this context and what opportunities does this offer for the future?
We see great potential here for the state of Berlin and for Berlin Recycling. What helps us all is to network more and learn from each other together. Especially with topics like innovation and AI, it is important to generate use cases together and share the results with each other, not every medium-sized company will be able to build up its own AI and innovation area, so networks are and will become all the more important.

Finally, based on your experience, what would you recommend to other companies considering using AI in their processes? Are there any specific lessons or insights you'd like to share?
Above all, it is important to inform all affected employees at an early stage and to take them along with you, to record and dispel any fears they may have. Communication is the be-all and end-all here. We found it particularly helpful to involve the works council in all decisions concerning the projects. This gave us a transparent and eye-to-eye relationship right from the start.

It is also important to find good partners who, on the one hand, are familiar with the subject of AI and, on the other, are able to explain it to lay people. Project planning with small work packages carried out together with the partners is also enormously important for success.. In addition, internal supporters are important, because only in this way can a common path be taken.

Thank you, Bianka Rieder, for your time.