Traffic Analysis Toolbox Volume X:
Localized Bottleneck Congestion Analysis
Focusing on What Analysis Tools Are Available, Necessary and
Productive for Localized Congestion Remediation

4.0 What Level of Analysis is Warranted?

An agency with unlimited resources could possibly study - and unnecessarily overanalyze - some proposed physical and operational improvements. Studies require time and cost. There have been many projects wherein low-cost congestion solutions were based on qualified engineering studies (perhaps only sketch planning was necessary) and were executed in the course of time savings and/or cost savings, without execution of a complicated simulated analysis. Conversely, modeling invariably provides a more detailed analysis. Agencies should consider the pros and cons of substituting sketch planning-level studies against a potentially marginal benefit of higher-level analysis. In short, the cost differential may be one factor but should not be the factor.

A clear understanding of both the study area and resources available for the project should fairly guide the decision of how much analysis is necessary.
There are two sides to this coin.

• The level of analysis should roughly correlate to the size of the problem. However...
• ... by their very nature, these are low-cost, "low hanging fruit" problems that nevertheless have potentially huge benefits in terms of reducing hours of delay. Even the simplest change may incur a significant operational change elsewhere on a facility.

An agency should not scrimp on the resources necessary to make a knowledgeable decision; meaning that if the complexity of the project, or the level of public discourse is inordinately high, the agency should prepare for a thorough analysis, and possibly a very involved public presentation and discussion, especially in locations near private entrances and land owners.

Some mitigation strategies do indeed lend themselves towards "obvious" solutions or even real world experimentation. Closing a low-volume driveway or ramp can always be reversed if necessary. Tweaking signal timing may indeed be cheaper than building a model and adjusting the simulated signal timing. On the other hand, structural changes to a freeway facility are too costly to rebuild or reverse; experimentation in these cases is not a wise option.

Another important consideration that must be made is the availability of good data. Often, no or little data are available, making a significant portion of the analysis cost devoted to collecting data. An agency also must weigh the cost and necessity of data collection as part of the overall cost, against the fidelity of the analysis results. The agency must include this consideration; is there sufficient data available for the level of study that this project warrants?

4.1 Project Guidelines

Every congestion mitigation strategy comes at a cost. This cost includes the hard costs of analysis, design, materials, and labor; and the soft costs of user impact, public opinion, and life-cycle costs. Building an overpass is an expensive operation. But mitigation strategies also carry sensory and learning costs too. Drivers who frequently traverse a specific corridor will have an adjustment period as they relearn their familiar route with the new overpass. All of these considerations must be weighed when considering to model. Ultimately, the agency's available resources, both in terms of preproject (e.g., analysis, public opinion, opportunity) and postproject (e.g., project cost, public acceptance, project life-cycle costs, interconnectivity, etc.) will decide how much project analysis is appropriate.

4.2 Small Corrections and Operational Changes

Many mitigation strategies have implementation and learning costs low enough to justify qualified engineering judgment in the real world. These include the following:

• Adjusting the timing of existing signals. The timings can be tweaked again, or reset to original settings.
• Placing new signs or signals. Signs can be tweaked either in message or relocation; removing or rebuilding signals is a significantly harder task (See "Note" below).
• Some aspects of lane restriping. Safety, above all else, should be addressed; but in essence, the striping could be tweaked.
• The installation of loop detectors. The detectors will not inherently disrupt traffic; only the application of their data will.

In cases where the bottlenecks are absolutely isolated from upstream and downstream influences, and the study area is small, it may be sufficient for the agency to commit only enough resources and decision necessary to implement the strategies listed above.

Note: In the context of the message above, bear in mind that signal optimization software exists that can model intersection operation at less expense than a full simulation analysis.

4.3 Large and Infrastructural Changes

Most projects have implications that are too high to risk without considering even the least amount of analysis, if only to concur, justify, or present findings in a manner that warrants a responsible decision. Most agencies would agree that simulation is a necessary step in larger project execution, as in these examples:

1. The construction of new facilities, such as auxiliary or mainline lanes and overpasses;
2. Complex movements, such as weaves, or the introduction of new movements; and
3. Any changes in required driver behavior, such as converting two-way roads to one-way roads, or other major redirections of flow, may be considered as "non-traditional" solutions requiring other levels of outreach and marketing with local officials and the motoring public.

As has been said before, these changes are expensive to implement, and would be prohibitively more expensive to undo or change. Agencies save money by spending resources up front to analyze simulations, and only implementing these major changes once.

4.4 Public Support and Justification

Of course, not every case is so clear. Planners and approving boards and councils are stewards of the public trust and budget. They must consider which strategies are appropriate for the area, and estimate the potential cost and impact of each strategy. A good rule of thumb is that the level of analysis should correlate to the perceived level of total mitigation cost. A computer-aided rendering of a before-and-after proposal may be its own justification to use micro simulation to present a proposed project to the public.