Maintenance pilot with Sis.Ter

Pilot in Delft

 

How can intervention rates can be optimised to maintain facilities?

 

QUICK FACTS

  • Project: Smart public waste management
  • Location: Delft, the Netherlands, with 103.000 inhabitants
  • Challenge: How can intervention rates be optimised to improve the public waste management in Delft?
  • Solution: A dashboard that uses data collected from sensors installed in public waste bins. This dashboard can be used as a forecasting tool to find out which areas will have the most waste bins that need to be emptied most frequently.

THE CHALLENGE

Using open data to optimise intervention rates in public waste management. Waste collectors usually work along a standard route to ensure that all public waste bins are frequently emptied. While this process works, it could be improved by ensuring that those collecting waste follow a route that allows them to empty especially those public bins that are (almost) full instead of emptying bins that contain no waste.

THE SOLUTION

Starting from open data provided by Delft’s municipality, the startup Sis.Ter initially developed a geographic decision support system (grid). This analysis has been done to define which areas of the city were most stressed by waste flows. This was the starting point to tackle the challenge, giving to the municipality an initial overview of critical areas of the city and waste production, according to different data and factors. For example, event locations and the estimated number of visitors. After this first output, the model was further improved by using data collected from 15 ultrasonic waste sensors that were installed in public bins. Data collected from sensors have been used to show real-time percentages of the filling grade on the determine the collection rates. Moreover, Sis.Ter developed a mobile application that can be used by public waste collectors, a responsive version of the dashboard itself.

 Fig, 1:  Thematic map with a special gradation that indicates areas of the city with greater criticality (red), stable ones (yellow) and “virtuous” ones (green) bins.

RESULTS & LEARNINGS

The main results from this pilot were the dashboard and mobile application. This give further insights into the areas most stressed by litter flows in the city of Delft. The dashboard confirmed that public bins in some areas collect much more waste than bins in other areas. This justifies the goal to adjust the routes of field operators to optimise intervention rates – as well as saving public money.

Due to limited time and resources, an application related to fleet management and collection’s tours optimisation was not developed.

One of the learnings was that half a year is quite short for the development of a specific mobile app for field operators as this app first required the creation of the dashboard itself and the collection of data over a longer period of time.

 

BENEFITS FOR OTHER CITIES

Cities that want to improve their public waste collection can certainly benefit from the pilot in Delft. Although the app for field operators has not been developed and the implementation still needs to follow, the created dashboard already provides a great example of how open data can be used to gather information on public waste.

The most interesting learning seems to be that it is not necessary to install sensors in all bins in the city to gain insights and optimising waste collection.

 

NEXT STEPS

There are on-going discussions on how to bring this pilot to the next phase of implementation. The solutions are being further tested by the city and its waste collector. There is contact between the city of Delft and the startup Sis.Ter and a final decision on whether and how this pilot will be implemented is expected to follow soon.