U·Sim – Assessment of Customer Usage Profiles

While success runs require the specification of sample size and test duration, larger measurement activities demand additionally the variation of influencing factors. Test engineers have to decide about the number of levels and the corresponding values the factors are adjusted to:

  • Road profile (city, country, motorway, off-road)
  • Load and kind of load (load fixed or tank truck)
  • Climate
  • Driving style (carefull, aggressive)
  • etc.

By the statistical methods of "Design of Experiments" (DoE) the design of the tests can be found fullfilling several optimality conditions, like minimum costs or minimum expenditure of time and material.

U·Sim Process

U·Sim process
© Photo ITWM

U·Sim process

Analysis of variance (ANOVA) and regression judge about the significance of the factors as well as their quantitive influence. Interaction effects (e.g. aggressive driver conducts a heavily loaded truck) may also be considered.

A probability distribution on the different classes of measurements describes the expected utilization of the vehicle for a specific group of customers. This enables us to recombine the measured data via Monte-Carlo simulation and estimate the distribution of quantities of interest for that group (service loads, fuel consumption, etc.).