The Impact of Working From Home on Travel Behaviour Quantifying the Impact of Key Variables on Working From Home in the Netherlands and Its Effect on Side Activities and Travel Behaviour Using a Bayesian Belief Network and LMS Integration
264 Downloads
Last Updated: 4-2026
This study examines the impact of working from home (WFH) on travel behaviour in the Netherlands. A review of twelve relevant and carefully chosen studies show that WFH is influenced by socio-demographic, job-related, and spatial characteristics, yet no prior research has analysed the Dutch context using revealed preference data combined with a non-linear approach that simultaneously considers intercorrelations between determinants and effects on travel behaviour. To address this gap, data from the 2016 Dutch Time Use Survey is used to construct a Bayesian Belief Network (BBN) in GeNIe Academic 5.0, with “Working Behaviour”, containing WFH, as the main node. Parent nodes represent variables affecting the extent of WFH, while child nodes capture side activities that increase on WFH days. Results show that job position, age, weekday, educational level, possession of a driving licence, gender, household composition, children’s ages, income, working hours, urban density, home ownership, household occupancy, distance to work, partner’s WFH, and housing type are the most influential determinants, ordered from most to least impactful. On WFH days, individuals tend to make more non-grocery shopping trips and engage more in hobbies, sports, grocery shopping, and other activities compared to days spent at the workplace. The BBN results were integrated into the Dutch national transport model (LMS) to model the spatial distribution of WFH. A comparison of four scenarios demonstrates that explicitly modelling WFH results in a more accurate spatial distribution, with higher prevalence in urban areas and among highly educated individuals. These findings highlight the importance of explicitly incorporating WFH into transport models and provide valuable input for policy development. Future research should include additional variables, such as job sector, to further improve explanatory power.
