Utilizing Wi-Fi Sensing and an Optimized Radius Algorithm to Count Passengers with Transfers to Enhance Bus Transit O-D Matrix

Abstract

The origin and destination (O-D) of public transit passengers are important for the planning and operation of the transit system. However, only 46% of public transit agencies have a smart card system in the US, and most of them require an entry-only tap, which prohibits identifying passenger destinations unless utilizing an estimation model. Therefore, there is a need for a cost-effective and automated solution to facilitate the majority of the US transit agencies in recognizing the origin and destination of passengers as well as capturing passenger transfers. This paper created a novel algorithm for transit agencies to count passengers with the consideration of transfers using a cost-effective Wi-Fi sensing–based approach. Two pilot studies were conducted in the city of Louisville, Kentucky on three different bus routes to explore the feasibility of the method. A Wi-Fi detector was installed in the bus to detect passengers, and a manual counting was performed to be used as ground truth data. After data collection, the proposed algorithms were applied to optimize the detection radius and to eventually find origins, destinations, and transfers. Analysis revealed that the proposed Wi-Fi–based approach is capable of recognizing 78.7% of the total passengers as well as detecting their boarding and alighting activities. The paper demonstrates the ability of the proposed method to detect passengers with a reasonable detection rate by using Wi-Fi technology on bus routes, which makes it feasible for transit agencies to conduct frequent and low-cost network-level passenger O-D studies. This work utilized Wi-Fi technology to count the passengers of public transit and to identify transfers. Current practice in data collection is transitioning from traditional manual methods toward more-automated approaches. The current practice relies on a manual method such as surveys or newer technologies such as gaining information from a smart card that passengers swipe when they get on a bus. Surveys are a manual method that requires intensive effort to gather the data and process it. On the other hand, the smart card system can be expensive to install and operate, and most public transit providers in the US have limited budgets and large coverage areas. The algorithm in this work is a practical solution for public transit agencies to gain information about the boarding and alighting activities along their routes. A Wi-Fi sensor installed inside the bus can be used to estimate the number of passengers who board and alight the bus. The results of the novel method in terms of detection rate demonstrated higher accuracy than previously proposed algorithms. This work also revealed the ability of the algorithm to capture transfer activities in terms of the number of passengers and other transfer characteristics.