On 27 May 2021 the webinar on cost-benefit analysis (CBA) was held via Microsoft Teams, which attrached 38 experts from around the world. People and institutions with interests in cost-benefit analysis of automotive safety systems were invited: vehicle safety, including road safety policy makers, vehicle manufacturers, road safety NGOs, vehicle assessment organisations, fleet owners and vehicle safety researchers.
The aim of the webinar was to present and explain the preliminary CBA tool and getting feedback from the experts on applicability of the tool.
In the first part the VIRTUAL project was presented, an introduction to CBA for automotive systems was given and the preliminary CBA tool was presented and explained (see slides for more information).
In the second part particpants were asked what they would use the tool for in their organisation. Most of the participants can imagine using the tool in research for example for injury/collsion reduction strategies for cyclists. The tool could be used in communication with policies and consumer rating.
The audience was also asked what developments are necessary in their view to improve the tool for their application. The consensus was that without having applied the tool, it is difficult to answer this question. At this stage there is no obvious component missing. However, a go-through example and predefined input values and/or data sources were mentioned as helpful.
Vice versa many questions were raised towards the developers of the tool. In focus were for example data sources and estimates for the tool. How is exposure/crash risk considered? Etc. The questions will help to make the tool more user friendly and easier to use.
Please find all questions and answers from this webinar below.
Questions from VIRTUAL towards participants
What would you use the CBA tool for in your organisation?
- We may use it in research activities, related to CBAs
- To evaluate socio economical benefits
- Research & Development
- Propose innovative products to management – decision if development of product is reasonable
- Could be for product viability and influence Euro NCAP
- Research regarding socio-economic-benefits
- Communication with regulatory policies & consumer rating
- The tool should be useful, e.g., also for PTW rider protection
- Compare different measures and decide which one we would suggest to our client
- Would like a CBA tool for injury/collision reduction strategies for cyclist collisions
Which further developments of the tool are necessary to improve the tool for your application?
- “Searching for input values might be challenging. Thus it could be practical to offer the user some set of predefined (default) input values.”
- “Not completely clear which limitations are there today. Any use cases which cannot be addressed today.”
- “An example with a go-through from crash risk source to final assessment. Sources for the calues to be input.”
Questions from participants
When will the VIRTUAL CBA-tool be available?
The final tool will be available at the end of the project in March 2022. The platform will continue after the project has ended.
Will the HBMs be publicly available?
Yes, VIRTUAL is based on an open source approach.
How were the use-cases for analysis selected?
We selected use-cases relevant for road safety, with focus on the link to the HBMs being developed. In VIRTUAL we collect data on a limited number of cases, but the CBA-tool is rather general and is applicable to other cases.
Who is the ideal user of the tool?
The tool requires knowledge of CBA, but the results of the tool are generally applicable for, for example, policymakers.
What is the target audience for the tool?
We believe that a combination between researchers and policy makers is the best way to go.
How can users find input values for injury probabilities?
It does require some efforts from users. For VIRTUAL use-cases these values are obtained from literature. Other sources include national road safety statistics in your country.
Any suggestions for method for determining risk percentages?
This is always a challenge. A large study of accident and injury reporting is beyond the scope of VIRTUAL. An example of European data is the Care database. Furthermore, the total number of vehicles in Europe is fairly well known and can be used as the denominator. Ideally there would be more detailed data, for instance, from larger naturalistic driving studies where vehicles are equipped with instruments that record everything. Preceding studies of this kind have been limited in scope. Any estimate that we can calculate will almost certainly be a minimum. Other possible sources include the in-depth German database GIDAS.
Could the product impact Euro NCAP?
Yes, VIRTUAL has a continuous dialogue with Euro NCAP.
Any use cases that cannot be addressed today?
The tool can address a basic set of use-cases. Of course, the goal is for the tool to encompass as much as possible.
Is it possible to provide default values for difficult to estimate input?
We will investigate if this is possible to provide
How is the tool taking exposure into consideration?
The most straightforward way is to use your own country’s statistics: recorded number of injuries or collision types divided by some estimate of number of vehicles or vehicle kilometers. Note that the tool can be used for sensitivity analysis.
It is very difficult to get all the input needed for the tool. It seems a lot of guesswork involved?
Yes, some of the inputs are difficult to estimate. In VIRTUAL we have partners in the vehicle industry that can help with estimations, but we are aware that this is a challenge for the users of the tool.
Is there a clear correlation between QALYS and MAIS?
Yes, MAIS measures injury severity and higher MAIS means higher QALYS. Note that QALYS and MAIS are still not widely used.