In this study, we aimed to evaluate the performance of various machine learning (ML) classifiers to predict mode choice of movement-challenged persons (MCPs) based on data collected through a questionnaire survey of 384 respondents in Dhaka, Bangladesh. The mode choice set consisted of CNG-driven auto-rickshaw, bus, walking, motorized rickshaw, and non-motorized rickshaw, which was found as the most prominent mode used by MCPs. Age, sex, income, travel time, and supporting instrument (as an indicator of the level of disability) utilized by MCPs were explored as predictive variables. Results from the different split ratios with 10-fold cross-validation were compared to evaluate model outcomes. A split ratio of 60% demonstrates the optimum accuracy. It was found that Multi-nominal Logistic Regression (MNL), K-Nearest Neighbors (KNN), and Linear Discriminant Analysis (LDA) show higher accuracy for the split ratio of 60%. Overfitting of bus and walking as a travel mode was found as a source of classification error. Travel time was identified as the most important factor influencing the selection of walking, CNG, and rickshaw for MNL, KNN, and LDA. LDA and KNN depict the supporting instrument as a more important factor in mode choice than MNL. The selection of rickshaw as a mode follows a relatively normal probability distribution, while probability distribution is negatively skewed for the other three modes.

Local government bodies and other concerned agencies in developing countries spend a considerable amount of money on rural road development. However, in the absence of any robust and systematic methodology, road development largely relies on ad-hoc decisions and subjective judgement of public officials. Such a decision-making process often leads to inefficient resource allocation bypassing equity and long-term societal benefits. Although there are some established methodologies for road network planning, complexities exist in applying those methods. First, most of the established methods are not suitable for rural road development, particularly regarding the volume and nature of traffic on them. Second, some methods are highly complex and lack practical applicability. Third, road development planning should not be top-down alone but ensure the participation of local stakeholders. Given these limitations, this study proposes a methodology—Rural Road Planning and Prioritisation Model (RPPM). It consists of two major components (i) developing a core network in participation with local stakeholders and (ii) prioritisation of roads based on Cost-Benefit Analysis (CBA) and Multi-Criteria Analysis (MCA). The proposed method is piloted in one district, and a web-based software is also developed for practical implementation by the Local Government Engineering Department (LGED), Bangladesh. The paper also discusses the results of the pilot study.