This report builds on an emerging body of literature that suggests capacity is not a constant value but is a variable. Capacity (or sustainable flow) depends on driving behaviour, the composition of the drivers at the moment, driver familiarity with the roadway, the mix of vehicle types, trip purposes, weather, signalization, and the presence of upstream or downstream bottlenecks. Therefore, if operational improvements are made to a roadway, the sustainable flow will be influenced by these variables. The research also suggests that the success of operational improvements in increasing sustainable flow is partially dependent on the configuration of the roadway network. The same effect is not necessarily achieved with every application. Finally, drivers learn and adapt to new roadway configurations. To properly analyse the effect of an operational improvement, this learning behaviour should be recognized. The implication of these contextual factors is that simulation methods are needed to estimate the increase in sustainable throughput achieved from a package of operational improvements. The researchers added capabilities to existing simulation models to reflect the stochastic nature of capacity for freeways and arterials and day-to-day learning in the route selection algorithms. They also used link, corridor, and network diagnostic features for evaluation of alternatives and to identify locations on the network where the benefits of implementing operational strategies appear high. The researchers evaluated emerging technologies and network operations treatments, from which they selected 25 for testing, including actions such as ramp metering, interchange modification, queue management, narrow lanes, adaptive signals, and pretrip information. A simulation modelling approach was used to test strategies on networks and data from Fort Worth, Texas, and Portland, Oregon. The results of applying the simulation techniques to the various strategies indicate the following: • Multiple performance measures are needed to obtain a complete assessment. • Measures should reflect link, corridor, and network characteristics. • Driver’s route choice must be considered to analyse network-level effects. • A representative cross section of corridors and O-D pairs should be evaluated. • The effectiveness of a particular operational improvement can only be evaluated in context. • The reliability of travel times may improve even when the actual travel time remains unchanged. The report describes the simulation modelling framework and the equations used for the enhancing simulation software. (Author/publisher)
Samenvatting