Loading…
D6.2 Analysis of coping capacity factors: vehicle and operator state
The current Deliverable aimed to provide the analysis results for the coping capacity factors, both for the vehicle as well as the operator state and the effect these have on risk. This aim was pursued by: (i) identifying the most critical factors of coping capacity, (ii) developing SEM and GLM mode...
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Default Report |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/25491682.v1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The current Deliverable aimed to provide the analysis results for the coping capacity factors, both for the vehicle as well as the operator state and the effect these have on risk. This aim was pursued by: (i) identifying the most critical factors of coping capacity, (ii) developing SEM and GLM model in order to investigate the effect of ‘vehicle state’ and ‘operator state’ on the STZ level and (iii) comparing the differences between different countries and transport modes. After making a short summary of the project’s aims and objective, the naturalistic driving experiment procedure in all of the countries involved was described along with the data acquisition, data cleaning and data aggregation procedures followed to extract the datasets that were used in the analyses. These strategies aimed to comprehend how the data were stored in the back-end database, how to deal with missing values, how to impute missing values taking into account the natural meaning of the recorded variables and how to best exploit the data for developing the Structural Equation Model (SEM). The volume, diversity and noise included in the dataset, due to the different experimental difficulties faced in each of the countries led to extensive efforts to acquire clean data. The next section of the Deliverable describes in detail, the methodologies followed throughout the analyses. Apart from SEMs, Generalized Linear Models (GLMs) were also used and the goodness-of-fit-metrics for the models were explained. The main results of those analyses are thoroughly described in Chapter 4 of the current Deliverable. The analysis found that age, confidence, and driving style were the strongest indicators for operator state, while vehicle age and gearbox were significant for vehicle state. Mixed results were found when looking at the correlation between coping capacity and risk in different countries and transport modes, however the majority of the modes point towards a negative correlation between coping capacity and risk (i.e. higher operator capacity leads to lower risk). The lack of objective coping capacity indicators in the study may have contributed to the lack of coherence between all the developed models over all countries and modes. However, there was consistency in the increase of coping capacity's effect on risk throughout the phases of the experiment. Despite efforts to clean and homogenize the data, an overall "coping capacity against risk" model for a specific mode was not possible due to the volume and diversity of the data. Future trials may provide additional data to help address these limitations and produce more conclusive results. Finally in the last chapter, conclusions are drawn for the relationship between coping capacity and risk, while explanations for the model drawbacks are given. |
|---|