"TimeSeries' experts were paramount in tying the complex landscape of technologies together."
- Alderik Bos - CEO Liftinzicht
To collect sensor data from sensors placed on elevators and analyse and process the data in a platform to predict the need for maintenance.
Connected elevators through the TimeSeries Analytics platform using IoT LoRa-connected sensors. Results were returned to Mendix to calculate and predict elevator maintenance schedules.
A user-centric app that predicts and allocates maintenance recources based on live data from the elevators.
Driving innovation in the fairly conservative elevator maintenance sector is a daring plan, but Liftinzicht accepted this challenge. Since this
decision was made, Liftinzicht has modernized and optimized elevator maintenance across the Netherlands. While partnering with industry leaders
like Kroon and Mitsubishi, they also chose to technically partner with TimeSeries to realise their innovative ambition.
In this case we will dive a bit deeper into a specific part of the overall Liftinzicht platform, predictive maintenance. So once the challenge was clear, immediately some questions arose around realizing this:
With these questions Liftinzicht and TimeSeries started the journey to solve this challenge.
We used the TimeSeries smart app value chain to discover what we needed to do, filling in activities for the Gather, Process, Inform and Act phases.
Link past maintenance and real time data to predict the elevator lifecycle.
Feed back information to planners to plan engineering schedules.
An engineer will be dispatched to
Once the smart app value chain was established, we needed to match the demand for gathering with the right techologies. For the actual sensor hardware we turned to a partner of Liftinzicht that manufactured the measurement devices. The next step was to get the data from those devices in elevator shafts to the TimeSeries Analytics platform in the Cloud. We used the LoRa Network to transfer the data to our platform. LoRa is a specialized network managed by KPN to enable IoT connectivity.
The left part of the screen displays an overview of the performance of the selected elevator trough the year. The systems combine information from planning and incident managament modules to map maintenance visits and disruption of service on the graph.
By determining the peak and off-peak hours of each elevator, maintenance can now be scheduled at times during which usage of the elevators is lowest, which in turn causes the least disruption to the users of these elevators. This is especially relevant in locations such as hospitals, where
the availability of an elevator is of critical importance and disruptions have to be minimized at all cost.
By applying our predictive maintenance algorithms to the gathered data we determine at which moment in the future maintenance is needed. This is a valuable alternative to the fixed maintenance plans which are usually used for elevators. Instead of maintaining each elevator in a building four times a year, maintenance can now scheduled based on the actual usage of these elevators.
The expectation is that the average number of saved maintenance jobs is 1,5 per elevator per year. For large building owners with over 800 elevators this leads to a cost reduction of € 180.000 per year.
The application of IoT technologies and the TimeSeries Analytics platform has led to a reduced cost of maintenance, fewer disruptions and a higher overall availability of the elevators.