01023224

Component

https://doi.org/10.3141/1935-19

http://scholar.google.com...agel&publication_year=2005

Q-learning is a method from artificial intelligence to solve the reinforcement learning problem (RLP), defined as follows. An agent is faced with a set of states, S. For each state s there is a set of actions, A(s), that the agent can take and that takes the agent (deterministically or stochastically) to another state. For each state the agent receives a (possibly stochastic) reward. The task is to select actions such that the reward is maximized. Activity generation is for demand generation in the context of transportation simulation. For each member of a synthetic population, a daily activity plan stating a sequence of activities (e.g., home-work-shop-home), including locations and times, needs to be found. Activities at different locations generate demand for transportation. Activity generation can be modeled as an RLP with the states given by the triple (type of activity, starting time of activity, time already spent at activity). The possible actions are either to stay at a given activity or to move to another activity. Rewards are given as “utility per time slice,” which corresponds to a coarse version of marginal utility. Q-learning has the property that, by repeating similar experiences over and over again, the agent looks forward in time; that is, the agent can also go on paths through state space in which high rewards are given only at the end. This paper presents computational results with such an algorithm for daily activity planning.

01023220

English

pp 163-169

2005

Transportation Research Record: Journal of the Transportation Research Board

0309094097

Print

Figures (1) ; References (13) ; Tables (6)

Algorithms; Machine learning; Travel demand

Activities of daily living; Daily activity plans; Marginal utility; Reinforcement learning

Highways; Planning and Forecasting; I72: Traffic and Transport Planning

TRIS, TRB, ATRI

Apr 25 2006 5:03PM

• Calibration of Microsimulation Models Using Nonparametric Statistical Techniques
• Development of Methodology to Design Advanced Traffic Surveillance Systems for Traffic Information Based on Origin–Destination
• Efficient Map Matching of Large Global Positioning System Data Sets: Tests on Speed-Monitoring Experiment in Zürich
• Evaluation of AASHTO SiteManager Performance on High-Latency Satellite Communication Links
• Exploring the Relationship Between Data Aggregation and Predictability to Provide Better Predictive Traffic Information
• Geographically Stratified Importance Sampling for the Calibration of Aggregated Destination Choice Models for Trip Distribution
• ISO 19133 Tracking and Navigation Standard: 6.6 Linear Reference System Standard
• Lane-Based Network for Flow Analysis and Inventory Management of Transportation Networks
• Markov Chain Monte Carlo Multiple Imputation Using Bayesian Networks for Incomplete Intelligent Transportation Systems Data
• Modeling Location Choice of Secondary Activities with a Social Network of Cooperative Agents
• Modeling Urban Link Travel Time with Inductive Loop Detector Data by Using the k-NN Method
• Robust Route Guidance Model Based on Advanced Traveler Information Systems
• Simple Map-Matching Algorithm Applied to Intelligent Winter Maintenance Vehicle Data
• Travel Time Prediction Algorithm Scalable to Freeway Networks with Many Nodes with Arbitrary Travel Routes
• Using a Fuzzy Quality Function Deployment Model to Identify Improvement Points in Airport Cargo Terminals
• Using Decision Support System and Graphical User Interface to Choose Appropriate Lane Configurations at Toll Facilities
• Using the Ant Algorithm to Derive Pareto Fronts for Multiobjective Siting of Emergency Service Facilities
• Vehicle Detection and Classification Using Model-Based and Fuzzy Logic Approaches