Strategies to Overcome Network Congestion in Infrastructure Systems
> download pdf of paper

by Jason W. Black and Richard C. Larson, Center for Engineering Systems Fundamentals, Engineering Systems Division. (submitted to IBM Systems Journal.)

November 2006

Introduction and Overview
Networked Infrastructure systems deliver services and/or products from point to point along the network. They include transportation networks (e.g., rails, highways, airports, sea ports), telecommunication networks (by frequency-bounded airwaves or cables), and utilities (e.g., electric power, water, gas, oil, sewage). Each is a fixed capacity system having marked time-of-day and day-of-week patterns of demand. Usually, the statistics of demand, including hourly patterns (i.e., means and variances) are well known and often correlated with outside factors such as weather (short term) and the general economy (longer term).

An infrastructure system is typically difficult and expensive to design and construct. Once built, it can have a mean lifetime from 20 years (telecommunications) to over 100 years (water). As population and the economy grow, increasingly large demands are being placed on infrastructure systems. Eventually they must be upgraded due to lack of adequate capacity and/or the need for improved technology. However, that moment can be delayed, often for long periods, by the use of congestion pricing to reduce peak demand. Congestion pricing provides incentives to shift demand from peak time periods to lower demand periods. This effectively increases the capacity of the system without the need for additional investment.

Current examples of congestion pricing schemes include: time of day congestion pricing for autos in Singapore and London; for-profit 'toll-ways' adjacent to freeways; time of day pricing for electricity; time of day pricing for long distance telephone calls; revenue management in airlines to balance out travel demands over the course of a week and over the year; and auction type bidding for some infrastructure services, with higher prices paid for congestion periods.

This paper investigates congestion pricing across critical infrastructures in terms of the potential benefits of forgone investment achieved by reducing peak demand. It also presents several existing implementations of congestion type pricing. We then look at the political and economic impediments to widespread adoption of such pricing schemes. Finally, the paper presents areas of future research to develop congestion pricing strategies that provide efficiency gains and are politically acceptable and amenable to implementation across infrastructure domains.