Are seniors in mountainous areas able to realize their desired trips? A novel approach to estimate trip demand

Abstract

Listen

Translate article

Public transport services in mountainous areas are limited, and seniors living in such areas experience various inconveniences in their daily lives. Therefore, the planning of adequate services is crucial in improving the quality of life. Conventionally, public transport services are planned based on observed trip demands, which are referred to as realized trip demands in this study. However, this approach is inadequate because seniors living in an area with insufficient public transport services are generally unable to realize all their desired trips. This study proposes a model to predict the desired trip demand based on the analysis of the travel conditions of seniors in mountainous areas. The proposed model predicts the desired trip demands based on both the realized and unrealized trip demands. The bivariate Poisson regression model is applied to the distribution of desired trip demands and mobility constraints, and the joint probability model, which considers the relationship between them, is proposed. Additionally, in the case where respondents are unable to report the unrealized trip demands accurately, an alternative method is implemented to estimate the desired trip demand based only on the information on whether unrealized trip demand exists, and its estimation results are discussed. The analysis results demonstrate that the conventional trip demand model, established using realized trip demands, underestimates the trip demand in the target area and show that the estimated parameters are biased. The conventional trip demand model neglects 30–50% of trip demands in the target area and cannot provide the trip demand required for appropriate transportation planning. The reproducibility of the desired trip demands can be further improved by extending the model to consider overdispersion or the case where unrealized trip demands have a large share of zero trips or both.