*story developed in partnership with TWAICE, battery health experts
A major operator of electric buses in the Netherlands reached out to us regarding our Battery Health Solution, offered in partnership with TWAICE. The operator in question is an early adopter of e-buses. They started to add these to their fleet in 2014 and had more than 100 electric buses by 2017. For a considerable number of buses, the five-year warranty period given by the OEM was about to expire. This raised several operational and strategic questions for the operator. The initial question, “What is the status of (each) battery actually?”, was the most pressing. This brought more questions with it:
- How do we get the SoH (State of Health) data?
- What are the concrete warranty terms for the specific buses?
- Were the assumptions made at the time of the purchasing decision valid?
- What is the aging behavior across the sub-fleet; did some buses age faster than others?
- What is the aging performance of buses from other OEMs; can a comparison between buses be made?
- What are the strategic and operational implications of the outcome?
- What are the lessons learned and how to incorporate them in the future?
As these questions were arising for the first time – on both the operator’s and likely the OEM’s side – no standard processes had been defined, which resulted in several challenges.
Trying to answer these questions presented both the operator and OEM with challenges of a very practical nature that, overall, resulted in a time-consuming and resource-intensive process. Much of this could have been avoided by using our Battery Health Package.
Prepare and organize tests: The operator did not have an overview of the current State of Health (SoH) within their bus fleet and after clarifying with their OEM, they learned that to provide this SoH, the OEM needed to run physical tests on the bus (charging and running the battery empty several times). The OEM and operator agreed to select a reference bus to conduct this relatively time-consuming procedure (the test took one week). Moreover, organizing this required planning and alignment between the operator and OEM. Many people were needed on both sides to make the decisions necessary to prepare for the test. Experts from the operator’s side spent more than 100 hours preparing for discussions with the OEM to understand warranty terms, the definitions of SoH and how to estimate it. Additionally, the physical test had implications for operations, as a bus had to be taken out of the pool for an entire week.
Reliability of test (one vehicle over one week): As a starting point, and to keep costs and impact reasonable, one bus was selected for testing. The OEM applied a standard approach to estimate battery SoH. At first glance, this may seem reasonable, but this approach has several shortcomings. Firstly, it relies on the State of Charge (SoC) signal and requires careful execution. Results are blurred if charging and discharging cycles are not executed with low C-rates (a measure of the rate at which a battery is being charged or discharged) and if the measurement interface is not chosen carefully. Secondly, buses are operated in different conditions (e.g., variation in routes); hence, the battery degradation for each bus is individual and it is neither advisable nor possible to extrapolate the results from one bus to the whole fleet of bus batteries. As the test outcome from the OEM then showed that the bus was below warranty criteria, this called for the testing of the whole fleet, causing alarm to the operator as the test procedure used, was cumbersome to scale.
Costs on both sides: To answer the – what should be simple – question of determining the SoH of batteries in order to evaluate warranty claims, both the operator and OEM needed to invest significant time and resources. The testing procedure required the OEM to have one driver full-time throughout one week and significant effort to coordinate and organize the testing. Judging by the effort needed to test only one bus, it was clear that requesting this procedure for all buses would not be bearable from neither the OEM’s nor the operator’s perspective. While it is hard to calculate the exact cost, we estimate that a minimum of EUR 20,000 would have been required for this procedure.
ViriCiti dashboard featuring the battery health estimations in the bottom-right corner
The combination of these challenges, the anticipated difficulties to roll this testing out to the whole fleet and the need to obtain a second opinion on their bus batteries’ SoH prompted the operator to use the Battery Health Package. Being an early ViriCiti customer, the operator was able to draw on a rich dataset of battery field data that is necessary for using the Battery Health Package. The approach provided some powerful solutions to the challenges at hand.
Fast without operational impacts: The solution includes a non-intrusive SoH estimation as it builds on the data integration with ViriCiti and on collected field data. As an independent third party, ViriCiti provides centralized data that is fully autonomous, transparent and OEM-agnostic. The combined systems of ViriCiti and TWAICE allow operators to obtain in-depth information on their vehicle batteries remotely. Thus, the Battery Health Package provides the SoH of a battery without the need to run physical tests, coordinate the testing between OEM and operator and ultimately, avoid impact on operations. In addition, being a purely data-driven analytics solution, the Battery Health Package was rolled out within six weeks across the whole fleet for every single vehicle. For a fleet of, e.g., 50 buses, this would have either taken a year, testing each bus after another, or massive parallel testing would have been required – taking out a whole fleet of buses for a week.
Reliable: The Battery Health analytics solution was quickly rolled out across the entire fleet to provide the necessary SoH for each bus. Thus, instead of sampling buses to test, the solution presents fleet-scale analysis of battery health and replaces guesswork with real data. After all, every bus (battery) degrades differently. In order to perform accurate battery health analyses, a big-data approach is crucial. ViriCiti’s telematics system stands out through analyzing large amounts of data with high accuracy, which in turn provides the most reliable ebus and battery insights currently on the market. Crucially, the analysis across buses revealed a highly accurate SoH against the end-of-life criterion – with results broken down on a per-bus-level including historic data.
Additionally, a tailored data analysis showed that the testing procedure applied by the OEM had potential drawbacks in terms of validity and was not detailed enough for the discussions on warranties. Overall, having third parties collecting and analyzing data both from OEMs and operators preempts possible or alleged biases, and can provide an agreed single source of truth.
Cost optimized: The Battery Health Package provided fast and reliable information about the SoH of each bus without any operational impacts. By implementing the solution across the whole fleet, significant time-, cost- and effort savings for both operator and OEM were realized compared to rolling out the original testing procedure across the whole fleet.
Domain expertise and foresight: ViriCiti, as a pioneer in electric mobility, pushed for a data driven approach in fleet management early on. Collecting significant amounts of in-depth data – down to battery cell level – is also key to a reliable battery health analysis. This foresight and the resulting data-availability present the basis for battery-driven use cases offered with TWAICE’s expertise in battery analytics software.
Data availability and integration: Non-intrusive estimation of battery health (SoH) was possible due to the broad range of data captured by ViriCiti in combination with the ViriCiti and TWAICE partnership. This integration enables a comprehensive and reliable view into the health of a battery, delivered via our Battery Health Package.
Visionary early adopters: Operators that understood the relevance of data for electric mobility fleet management even when it was not obvious are another success factor for delivering the data driven Battery Health Package. Embedding a data-driven strategy into fleet operations from the start helped to have data readily available and enabled speedy processing.
Trusted & proven partnerships: The close and trustful relationship between all three parties involved (ViriCiti, TWAICE & operator) ensured a timely delivery of the solution.
Would you like to see an example of a battery health and lifetime prediction?