BikeSquare SmartBikes Predictive Maintenance
Transferring technological solutions to e-bike producers and renters to achieve predictive maintenance through AI and ML.
BikeSquare is an innovative startup with a social vocation that has developed high expertise in the cycling sector and has defined a model of cycle tourism circuit that contemplates the networking of public and private entities. The company has defined guidelines for the realization of works and regulations for the adhesion of private entities. This organization is on the market with four specific products: Platform web WhiteLabel + APP, Regional Portals (ex: http://www.cicloturismo.piemonte.it) and APP, The e-bike cycling portal (http://ebike.bikesquare.eu), and a Smart-Bikes Platform: Fleet Management, Analytics, Administrative Console.
The last one of the products is mainly related to operators in the sector (Renters, Tour Operators, Tourist Promotion Consortia, etc.) who wish to have a complete platform for the management of their route or their rental services. This system allows knowing where the bikes are located, turning on or off the bikes remotely if the user has already paid (or not) for the use of the bike, monitoring the bikes in case of theft, counting the commissions between the various partners, identifying the most popular routes, the best structures, etc.
This product consists of a web page platform and an IoT device for geolocation (optional) that the renter installs on the e-bikes. Tour operators can offer self-guided tours to their customers and get the visibility they need on the Web and APP. With the information collected with the IoT devices and the web page platform, we have a lot of information to implement different AI algorithms. For this pilot, we would like to implement a predictive maintenance system for the e-bikes and an alert system that gives us the status of the bicycles and IoT devices.
Predictive maintenance is a novel research topic that not only allows e-bike users to ride bicycles safely but helps e-bike renters to save money and time. For example, if the system suggests the company to buy new brakes in advance, the company will change them on time avoiding damages and having non-operative bikes while waiting for the brakes to arrive. Therefore, the company will prevent money waste.
Regarding the alert system, BikeSquare manages many e-bikes. Therefore, it is hard to know the status of all of them. For that reason, it could be helpful to have a system that notifies if a bicycle that disappears, moves unexpectedly, or does not send the geolocation information for a very long time.
- Organize and add the necessary information to perform the pilot.
- Search and select the correct algorithm that was able to perform in a proper way the prediction of possible damages or failures in the electric bikes.
- Search and select the correct algorithm that was able to perform in a proper way the detection of abnormal behavior in the rented bicycles.
- Verify that the technical infrastructure is able to execute the selected algorithms.
- Find the optimal evaluation methods in order to determine the best results.
The first milestone is defined after the three first months of the execution of the pilot. At this time the first objective should be covered , which is the development and the evaluation of the predictive maintenance algorithm using the services provided by EOSC.
The second milestone corresponds to the following three months. The team should finish the third objective which consists of developing the alert system.
The final milestone corresponds to the last three months. In which the engineers have to perform the last two objectives that consist of evaluating the development algorithms and construct a solid economic solution for the company.