Performance Assessment of Urban Street Facilities Addressing Improvement Issues of Automobile Mode

Abstract

Over the generations, automobiles mode is progressively reshaping patterns of living by increasing the accessibility and interconnectedness between different regions. Hence, sustainability of community development largely depends upon the establishment of efficient transportation network to serve automobile mode effectively in an urban corridor. The review of literature clarifies that; the service quality of urban street facilities has been evaluated since nineties. However, urban street infrastructural developments in emerging countries like India is still far away from satisfactory irrespective of current improvements based on the applications of existing evaluation strategies. This is due to the fact that: application of LOS model developed for homogeneous traffic conditions will fail to quantify the service provision criteria and heterogeneity of urban streets in developing countries. With the aim of justifying the existing approaches, this study proposed suitable Automobile Level of Service (ALOS) models to assess service quality provided by three different urban street facilities, namely, road links, signalised intersections and uncontrolled intersections operating under heterogeneous traffic flow conditions.
To achieve the objective of this research, a broad spectrum of geometrical, traffic operational, built-environmental and behavioural data sets are collected from a widely varying driving environment through field investigations, videography techniques, and perception survey. About 102 road links, 45 signalised intersections and 47 unsignalized intersections from nine Indian cities, are investigated in this study. Each investigated site has been rated on a scale of 1–6 (worst–excellent) by at least 20 drivers immediately after driving. The influence of different service attributes towards perceived ALOS score is examined with the help of Spearman’s correlation analysis. Six significant variables are identified to have influence on drivers’ satisfaction level at urban road links, such as: Peak hour traffic volume per effective road width (PHV/ We), Average travel speed (Savg), Pavement condition index (PCI), On-street parking turnover (P), Land use pattern (LU) and Hindrance factors (HF). Similarly, PHV/ We, Control delay (d), PCI, Queue length (Ql), Interruptions due to non-motorized traffic (ONV) and oppositely moving Encounters (OE) are the variables, which have significant influence on drivers’ riding quality at signalized intersections. For uncontrolled intersections, PCI, Presence of median barrier (PoM), Critical gap (tc), LU, OE and Service delay (Ts) are found to have significant influence on driver’s comfort level under heterogeneous traffic flow conditions.

Resulting significant variables are used as input variables in different statistical techniques and Artificial Intelligence techniques, namely, ordered logit modelling, ordered probit modelling, Artificial Neural Network (ANN), Functionally Linked Artificial Neural Network (FLANN) and Differential Evolution (DE), to predict ALOS scores of respective street facilities. Prediction performance of different modelling approaches are tested based on several statistical parameters and FLANN model is found to be the most efficient one in defining ALOS scores of urban road links with coefficient of determination (R2) value of 0.934 and 0.908 for training and testing datasets respectively. FLANN model is also found out to be suitable in defining ALOS scores of signalized intersections with R2 values of 0.949 and 0.895 for training and testing datasets respectively. Whereas, Bayesian Regularization Neural Network is found to be suitable in defining ALOS scores of uncontrolled intersections with R2 of 0.947 and 0.928 for training and testing datasets respectively. An associative ALOS score is assigned to six classes (A-F). More than 70% of studied road facilities are examined to offer service categories of ‘C’ or below.
The relative importance of each service attribute is determined with the help of model sensitivity analysis, and ranked in order of specific percentage value based on their degree of relative importance. PCI is reported to have the influence on drivers’ riding quality at road links and uncontrolled intersections with relative percentage contribution of 26.783% and 37.116% respectively. Hence, the pavement surface requires regular maintenance for smooth ride of vehicles, which in turn increase the serviceability to achieve a higher ALOS category. Whereas, Ql plays a significant role in fixing ALOS standards of signalized intersections, which have highest negative influence of 67.153%. Hence, optimizing traffic signalization timings and increasing effective green time for major approaches of an intersection in peak hours will significantly enhance the service quality of respective intersections. Similarly, other parameters are ranked in decreasing order of their relative importance. Based on the sensitivity analysis report, several improvement strategies are outlined in this study, which will help the transport authorities to identify operational issues of existing street facilities, and to design a users’ friendly transport system with better driving environment.