Travel behaviour of cemetery and columbarium visitors : models and policy implications

Abstract

Grave-sweeping refers to the activity of visiting cemetery and columbaria, which is one of the traditional cultures in Asian countries. The visits are concentrated in the two traditional grave-sweeping festivals every year. The high travel demands have caused traffic problems such as overcrowding and congestion. In Hong Kong, as the government had planned to build new columbaria in response to the great demand of cremation in the coming decade, the traffic problems will be worse if the current transport policies do not improve to take it into account. To improve the policies, it is necessary to consider and model the travel behaviour of cemetery and columbarium visitors. Although there were studies conducted for other special events, their models could not be applied to grave-sweeping due to the substantial difference in the travel characteristics between grave-sweeping and other special events. This study is therefore initiated to model the travel behaviour of cemetery and columbarium visitors.

This thesis models the travel behaviour in three steps: trip attraction, trip distribution, and arrival time period and mode choice. The trip attraction is modelled by enhancing the traditional trip rate model into a non-linear regression model. The trip attraction model captures the variation of trip rates attracted to cemeteries and columbaria with a date factor. The trip distribution is modelled by a multinomial logit model and captures the visitors’ choice of interchange locations of their trips to and from cemeteries and/or columbaria. Finally, a joint logit model is developed to model the visitors’ choice of the arrival time period in a day and the mode between the interchange location and the site.

The models developed provide a valuable reference for the government and the police to plan for the public transport provision and to formulate appropriate and effective traffic management measures respectively. The trip attraction result shows that the travel demand is peaked at the festival date and decreases exponentially with the number of dates deviated from the festival date. The trip distribution result reveals that the visitors generally prefer making an interchange at a location near to the columbarium, with a shopping centre and a direct public transport service available to the site. The arrival time period and mode choice result demonstrates that visitors have resistance in changing their arrival time period. Providing discount in travel fare or more frequent service to reduce the waiting time during the alternative time periods could provide incentive to visitors to change their arrival time period. Policy implications are given for the government and police to evaluate the current transport policies and to plan for the transport policies in the future to prevent the traffic problems from repeating.