Optimal Peak-Load Pricing, Investment, and Service Levels on Urban Expressways
Theodore E. Keeler, Kenneth A. Small
Abstract
Theodore E. Keeler, Kenneth A. Small
Abstract
Optimal tolls, capacities, and service levels for highways can be determined jointly by way of an integrated peak-load pricing model. In this paper, such a model is developed and estimated with data for roads in the San Francisco Bay Area. The results suggest optimal peak user tolls of 2-7 cents per automobile mile on rural highways, 2-9 cents on suburban highways, and 6-35 cents on central city highways. Although our results are to some degree dependent on the interest rate, time value, and peak demand configuration assumed, one basic conclusion holds up under all alternative assumptions: current user charges are well below optimal peak tolls. However, our results also suggest considerably higher rush-hour speeds than currently prevail on Bay Area roads, and the lower travel time costs suggested by our analysis (relative to the current situation) should to some degree offset the corresponding higher user charges.
OpenAlex reports 267 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Optimal tolls, capacities, and service levels for highways can be determined jointly by way of an integrated peak-load pricing model. In this paper, such a model is developed and estimated with data for roads in the San Francisco Bay Area. The results suggest optimal peak user tolls of 2-7 cents per automobile mile on rural highways, 2-9 cents on suburban highways, and 6-35 cents on central city highways. Although our results are to some degree dependent on the interest rate, time value, and peak demand configuration assumed, one basic conclusion holds up under all alternative assumptions: current user charges are well below optimal peak tolls. However, our results also suggest considerably higher rush-hour speeds than currently prevail on Bay Area roads, and the lower travel time costs suggested by our analysis (relative to the current situation) should to some degree offset the corresponding higher user charges.
Key concepts: Offset (computer science), Mile, Transport engineering, Bay, Investment (military), Value of time, Service (business), Value (mathematics)