Development of an Inference System for Drivers’ Driving Style and Workload Sensitivity from their Demographic Characteristics

Sato, Toshihisa; Akamatsu, Motoyuki · 2022 · Crossref

DOI: 10.54941/ahfe100737

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the need for personalized driver assistance systems by investigating the relationships between drivers’ demographic characteristics, their driving styles, and their workload sensitivity. Current assistance systems, such as Adaptive Cruise Control and Lane Departure Warning, utilize uniform specifications that do not account for individual differences in driver acceptance or cognitive load. The authors hypothesize that demographic factors influence static driving behaviors and sensitivity to workload, necessitating a method to infer these traits from basic demographic data to tailor system interfaces effectively. To test this hypothesis, the researchers conducted a large-scale questionnaire survey involving 1,616 drivers across Japan. The study excluded professional drivers and those with minimal driving experience. Data collected included demographic variables: age, gender, driving experience, annual mileage, driving frequency, and region. Driving style was assessed using the Driving Style Questionnaire (DSQ), which measures eight factors including self-confidence, hesitation, impatience, and anxiety. Workload sensitivity was evaluated using the Workload Sensitivity Questionnaire (WSQ), covering nine factors such as understanding traffic situations, physical fatigue, and concentration disturbance. The authors applied Bayesian network modeling to analyze the conditional probabilities between demographic characteristics and the DSQ/WSQ factors. The model structure was determined using the K-2 algorithm with the Akaike Information Criterion (AIC) for assessment, assuming demographic variables as parent nodes without interdependencies among themselves. The results indicate that "gender" is the most influential demographic factor, affecting more DSQ and WSQ factors than any other characteristic. Conversely, the model revealed no significant influence of "driving frequency" on either driving style or workload sensitivity, and no influence of "driving experience" on workload sensitivity. The estimated Bayesian networks successfully modeled the distributions of DSQ and WSQ scores based on demographic inputs. This allows for the inference of an individual driver’s likely driving style and workload sensitivity profile solely from their demographic data. The significance of this work lies in the development of an inference system that can predict individual driver characteristics without requiring extensive prior behavioral data. By establishing conditional probability distributions for driving styles and workload sensitivities across different demographic groups, the system enables the evaluation of a specific driver’s scores relative to their demographic peers. This capability facilitates the selection of target drivers for specific assistance system designs and supports the development of next-generation, adaptive driver assistance systems that adjust warning timings and control specifications to individual user profiles, thereby potentially improving safety and user acceptance.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• stress driving
• cognitive capacity variation
• sex gender
• mental demand
• older drivers
• personality driving

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data
• Theoretical Contribution: theory or model, computational model