Optimal Heterogeneous Sensor Deployment Strategy for Dynamic Origin–Destination Demand Estimation

Abstract

A network sensor location problem (NSLP) model is proposed to determine the optimal heterogeneous sensor deployment strategy in terms of the number of links (counting) and node (video or image) sensors and their installation locations for dynamic origin–destination (O-D) demand estimates under a budget constraint. The proposed NSLP model also captures the impact of the time duration for which traffic data measurements are available on the optimal sensor deployment strategy; this topic had not been addressed in previous studies. A sequential sensor location algorithm that avoids matrix inversions was proposed to solve the NSLP model under the assumption of multivariate normal distribution for the prior dynamic O-D demand estimates. A network from a part of Chennai, India, was used to illustrate the performance of the proposed NSLP model and solution algorithm. The results show that the proposed algorithm can identify the sensor locations on the basis of the amount of information provided and can determine the number of each sensor type to maximize the accuracy of O-D demand estimation. The results also illustrate that the proposed NSLP model can capture the time-varying characteristics of the amount of information provided by each sensor location. The proposed method can be used to analyze the impacts of various sensor types, their number, location, and the time duration for which traffic data measurements are available for estimation of dynamic O-D demand under limited resources.