Gupta, N., Gottapu, S.K., Nayak, R., Gupta, A.K., Derawi, M., Khakurel, J. (Eds.). (2022). Human-Machine Interaction and IoT Applications for a Smarter World (1st ed.). CRC Press.
Abstract:
A wide range of challenges to the passengers on safety, security, comfort, and reduced waiting times are the main aspects to be considered in first/last-mile services. With the advent of autonomous vehicles (AVs), first/last-mile transits can be more safe, reliable, and sustainable in the transportation sector. Developing a system, which handles these constraints and is equipped with state-of-the-art support, is always a challenging one. Moreover, it is noticed that the passengers are bound to opt for other transit such as the private modes over public modes of transportation.
This chapter proposes a technique to solve first-mile (FM) ridesharing problems with AV as a shuttle service and uses an Internet-of-Things (IoT) platform. FM transit refers to transfer of the passengers from their pick-up locations to the nearest public transport hub. The FM transit is solved by dividing the problem into two sub-problems. First, the pick-up locations are associated with the neighboring vehicles available. Here, we propose two techniques based on mixed-integer linear programming (MILP) and http://www.w3.org/1998/Math/MathML"> K https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003268796/ff7c0afb-bead-450a-a822-68dbcc013133/content/math11_1.tif" xmlns:xlink="http://www.w3.org/1999/xlink"/> -means clustering, in which the passenger requests and the vehicles’ data are collected in an IoT cloud. The passengers are allowed to submit travel requests through a human-machine interface (HMI), such as cell phones or automatic booking kiosks within a period. At the backend, an IoT cloud computes appropriate vehicles to serve the passengers at their respective pick-up locations based on the nearest location. Thus, the pick-up locations are clustered for different vehicles in the service network on the IoT cloud.
Second, routing AV to the public transport hub by minimizing the total distance traveled is performed. Here, we present a traveling salesmen problem (TSP)-based solution to route each AV more efficiently. The final route for the vehicles to follow is computed in the IoT cloud and then it is communicated to the respective vehicle. The routes of the vehicles can be changed dynamically while serving the current passengers based on the condition that the new request does not change the current route drastically. The new incoming requests are dynamically allocated to the nearest vehicle from the IoT cloud and a detour from the current route is suggested if the cost of including the new pick-up location is accountable.
This chapter focuses on each of the two techniques proposed for associating pick-up locations and is used with the traveling salesmen-based routing technique that showcases the overall solution to the FM ridesharing problem. Various numerical experiments are conducted for the proposed solution methodology using different combinations of the number of pick-up locations and vehicles. Our results show that the usage of the MILP approach proved to be more efficient for smaller fleet sizes. Although both methods performed equally for larger fleets too, the total vehicle miles traveled (VMT) for the entire fleet is lesser for the MILP approach when compared to the http://www.w3.org/1998/Math/MathML"> K https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781003268796/ff7c0afb-bead-450a-a822-68dbcc013133/content/math11_2.tif" xmlns:xlink="http://www.w3.org/1999/xlink"/> -means approach.
Furthermore, the proposed techniques ensure more request services until the total vehicle capacity gets exhausted. Therefore, this approach encourages the passengers to use public transport for their FM transit
License type:
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Funding Info:
This research / project is supported by the National Research Foundation - Virtual Singapore Program
Grant Reference no. : NRF2017VSG-AT3DCM001-018
Description:
This is an Accepted Manuscript of a book chapter published by CRC Press in Human-Machine Interaction and IoT Applications for a Smarter World on 2nd August 2022, available online:
https://www.taylorfrancis.com/chapters/edit/10.1201/9781003268796-15/first-mile-ridesharing-using-autonomous-shuttle-service-iot-cloud-platform-shyam-sundar-rampalli-pranjal-vyas-anuj-abraham-justin-dauwels?context=ubx&refId=9c9a4d58-b797-4562-9e66-8abd739a5551