From Traditional to Automated Mobility on Demand: A Comprehensive Framework for Modeling On-Demand Services in SimMobility

Abstract

Mobility on demand (MoD) systems have recently emerged as a promising paradigm for sustainable personal urban mobility in cities. In the context of multi-agent simulation technology, the state-of-the-art lacks a platform that captures the dynamics between decentralized driver decision-making and the centralized coordinated decision-making. This work aims to fill this gap by introducing a comprehensive framework that models various facets of MoD, namely heterogeneous MoD driver decision-making and coordinated fleet management within SimMobility, an agent- and activity-based demand model integrated with a dynamic multi-modal network assignment model. To facilitate such a study, we propose an event-based modeling framework. Behavioral models were estimated to characterize the decision-making of drivers using a GPS dataset from a major MoD fleet operator in Singapore. The proposed framework was designed to accommodate behaviors of multiple on-demand services such as traditional MoD, Lyft-like services, and automated MoD (AMoD) services which interact with traffic simulators and a multi-modal transportation network. We demonstrate the benefits of the proposed framework through a large-scale case study in Singapore comparing the fully decentralized traditional MoD with the future AMoD services in a realistic simulation setting. We found that AMoD results in a more efficient service even with increased demand. Parking strategies and fleet sizes will also have an effect on user satisfaction and network performance.

Integrating Supply and Demand Perspectives for a Large-Scale Simulation of Shared Autonomous Vehicles

“…A constant fleet assumption modeled here can be improved using time-dependent fleet sizing based on dynamic pricing. Similarly, a thorough TNC demand model needs to include the human tendency for cruising similar to ( 28 ). Once pooling is modeled, a choice dimension for choosing to pool can also improve realism and is ongoing work.…”

“…This France-based study showed that incorporating the demand-side model considerably affected the demographic for SAV use bringing more realism to the model. SimMobility is able to integrate supply and demand fully, and was used to compare current TNC operation with an SAV fleet for Singapore ( 27 , 28 ), with results showcasing a benefit from centralized control. However, no study has been able to integrate supply and demand while being able to study a 24-h simulation at scale (i.e., full population).…”

A validated multi-agent simulation test bed to evaluate congestion pricing policies on population segments by time-of-day in New York City

“…There are several well-known ABMS platforms designed to support decision-making, including but not limited to Transportation Analysis and Simulation System (TRANSIMS) (Nagel et al, 1999), MATSim (Balmer et al, 2009), Sacramento Activity-Based Travel Demand Simulation Model (SACSIM) (Bradley et al, 2012) Simulator of Activities, Greenhouse Emissions, Networks, and Travel (SimAGENT) (Goulias et al, 2011), Polaris (Auld et al, 2016), SimMobility (e.g. Nahmias-Biran et al, 2019), etc. TRANSIMS was a first-generation tool developed by the Federal Highway Administration (FHWA), after which the creators used to produce the next generation tool MATSim.…”

A Simulation-Based Evaluation of a Cargo-Hitching Service for E-Commerce Using Mobility-on-Demand Vehicles

“…For the matching of requests to vehicles, an insertion heuristic is used, which maintains a schedule for each vehicle and attempts to insert incoming requests into the existing schedules of nearby vehicles within a pre-specified search radius equal to 5 km, to ensure that waiting times and travel times of all passengers are within pre-defined thresholds (15 min for maximum wait/travel times). More details of the matching heuristic may be found in [39,41,42]. Minor modifications are made to the matching heuristic to model the cargo-hitching service.…”

“…The pre-day demand model for 2030 relies on a calibrated, activity-based model system for the year 2012 (that matches observed tour/stop generation rates, activity shares, and modes closely) estimated using household travel survey data [33,39]. This model was enhanced with the MOD modes by assuming a similar utility specification of taxis and by calibrating relevant alternative specific constants and scale parameters of the mode and mode-destination choice models against aggregate data on the usage of single-ride and shared-ride MOD modes [42]. The calibration and validation also included matching simulated outputs to observed screen-line counts, public transit smart card data, and network travel times (for more details on the model calibration, the reader is referred to Oh et al [39]).…”