A core objective of the ICM system is to help corridor stakeholders respond to traffic incidents in a fully coordinated, corridor-centric way to route travelers safely around the scene with minimal disruption and delay. What is the best way to do this, and what response plans should be in place to make it happen? These are central question for the I-210 Pilot, and the Connected Corridors team and project stakeholders have worked to:
- map out possible alternate routes for response plans
- identify necessary field elements to include in the plans
- determine how to communicate with agency personnel and travelers about the plans
- develop the rules and rules engine needed for the Decision Support System to evaluate and recommend response plans
- design the response plan deployment process
Identifying alternate routes
One of the first steps in defining response plans was to identify possible alternate routes in the corridor that could be used to reroute traffic around incidents. Working with stakeholders, the project team identified about 300 preliminary alternate routes. After several iterations to refine the number of options and match them with available ITS elements, the list was narrowed down to about 60 alternate routes for both east and west directions. The intent is to provide 1-3 alternate routes for any incident location.
The following graphic shows the beginning stages of identifying possible alternate routes to be used in the response plans:
Including other components
In addition to alternate routes, response plans include a range of other components, such as:
- signal and metering light timing changes to facilitate reroutes
- requests for equipment and personnel needed to implement the response plan
- plans for communicating with equipment, personnel, and travelers involved in or affected by the response plan
Project stakeholders and the Connected Corridors team have worked to identify these components and to define when and where they will be deployed as part of a response plan.
Rules and rules engine
The response plan elements and best practices identified by project stakeholders—the alternate routes to be considered, the actions to be taken, the components to be activated, the personnel to be involved—are captured as rules to be used by the Decision Support System in generating response plans. These rules, developed by the project team, take the form of: “If an event occurs at location X, do Y and notify Z.”
The rules will be processed by a rules engine, which will evaluate many rules and determine the best response plans. The rules engine must support a complex set of conditions to account for the complexity of incidents, considering factors such as time of day, day of the week, expected duration, location of the incident, and possible issues with alternate routes such as an active school zone. A response plan can also be scaled depending on the severity and length of the incident. For example, if two lanes of the freeway are expected to be closed for one hour, that would warrant a different response plan than if the two lanes were expected to be closed for ten hours or if the closure were to occur at 3am versus 3pm.
Deploying response plans
The overall process for deploying a response plan is:
- First, an incident must be identified and characterized with information such as its location, severity, and anticipated clearance time. Most of the time this will likely be done by Caltrans TMC operators, as they monitor the corridor 24/7.
- This information will then be passed to the Decision Support System (DSS) which will use the rules engine to suggest one or more response plans.
- Simulation models will then be used by the DSS to rank the possible response plans based on how much improvement each is likely to provide.
- Depending on the situation, the top response plan could then be automatically implemented or reviewed by transportation management staff who could either implement, modify, or reject the response plan.
This process will take approximately 10 minutes from start to finish. The following graphic illustrates the steps: