01456607

Component

http://www.trb.org/Publications/Blurbs/168134.aspx

Any decision regarding the design and construction of transportation infrastructure should be prompt and linked to the entire transportation system infrastructure. Transportation networks typically span a large geographical area and have a modular structure that consists of many subsystems with actors and sensors, while involving both continuous and discrete dynamics that evolve over different time scales. The difficulty in such a case arises from the fact that most often the available information is quite intensive; in a network-level thinking, a large amount of transportation assets should be taken into consideration that, many times, may impose conflicting design and construction goals. In transportation infrastructure design and construction problems there is the need to jointly consider information that may be measured and quantified and other information that is subjective and should be qualitatively assessed. Artificial intelligence (AI) provides methodologies for developing flexible multivariate models for solving difficult approximation and optimization problems. The flexibility of AI methods [e.g., knowledge-based systems, expert systems, pattern recognition, machine learning, neural networks, genetic algorithms (GAs) and evolutionary computation, fuzzy systems, etc.], as well as their performance in various interdisciplinary applications are well suited for the complexity and variety of transportation systems. AI has been applied to various fields of transportation engineering. AI has also played a critical role in solving many infrastructure problems related to design and construction (including maintenance). Some of the AI techniques that are most popular in infrastructure design and construction are GAs, simulated annealing, ant algorithms, and fuzzy logic.

01456594

English

pp 121-133

2012-11

Transportation Research Circular

Web

Figures; References; Tables

Artificial intelligence; Asset management; Construction; Design; Expert systems; Fuzzy logic; Fuzzy systems; Genetic algorithms; Infrastructure; Machine learning; Network analysis (Planning); Neural networks; Pattern recognition systems; Simulated annealing; Transportation; Transportation engineering

Evolutionary computation

Construction; Data and Information Technology; Design; Transportation (General); I20: Design and Planning of Transport Infrastructure; I50: Construction and Supervision of Construction

TRIS, TRB, ATRI

Dec 11 2012 9:58AM

• Advantages and Limitations of Artificial Intelligence
• Agent-Based Modeling and Simulation
• Analysis of Public Transportation Planning and Operations
• Artificial Intelligence and Microscopic Traffic Simulation Models: Applications to Parameter Calibration and Origin–Destination Estimation
• Difference Between Artificial Intelligence and Traditional Methods
• Neural Networks for Travel Time Prediction on Interrupted Flow Facilities
• Ramp Metering
• Short-Term Traffic and Travel Time Prediction Models
• Traffic Signal Timing and Optimization
• Transportation Safety Analysis
• Transportation Security
• Travel Behavior Research
• Urban Travel Demand Forecasting