As of October 1, 2016, the SWUTC concluded its 28 years of operation and is no longer an active center of the Texas A&M Transportation Institute. The archived SWUTC website remains available here.

600451-00008

SWUTC Research Project Description

Decision Theory Models for Selecting Traffic Control Devices

University:  Texas A&M University

Principal Investigator:
Gene Hawkins
Texas Transportation Institute
(979) 845-9946

Project Monitor:
Tom Hicks
Director of Traffic & Safety
Maryland State Highway Admnistration
7491 Connelley Drive
Hanover, MD 21076

Funding Source:  USDOT and State of Texas General Revenue Funds

Total Project Cost: $63,600

Project Number:  600451-00008

Date Started: 4/1/12

Estimated Completion Date:  3/31/13

Project Summary

Project Abstract:
The Manual on Uniform Traffic Control Devices (MUTCD) documents standards, guidance, and options related to the selection of traffic control devices. While the MUTCD provides support statements for some traffic control devices, it is beyond the manual’s scope to provide the depth of knowledge needed to evaluate traffic control devices as part of the larger transportation system; this is the purpose of engineering judgment within the manual. However, transportation professionals with limited experience would benefit from this additional information being available in a well-structure document; presently, outside the MUTCD, there is no well structured document on traffic control device selection matching the caliber of the Highway Safety Manual or Highway Capacity Manual. Additionally, research has yet to document the fundamental principles for a well-structured document on traffic control device selection. Therefore, using statistically valid descriptive and normative decision theory models, this research will develop the fundamental principles for a well-structured document on selecting traffic control devices, use these principles to evaluate a specific roadway situation (e.g. urban arterials), and provide a method for using the fundamental principles to develop decision theory models for other roadway situations.

Project Objectives:
The objectives of this research are to:

  • Define the selection of traffic control devices as a decision theory problem.
  • Collect historical data on traffic control device selection for a selected roadway situation.
  • Define objectives for selecting traffic control devices for a selected roadway situation.
  • Develop and conduct a survey of professionals investigating the way professionals with limited experience select traffic control devices compared to area experts.
  • Develop statistically valid descriptive decision theory models using historical data and survey results
  • Develop nominal decision theory models for selecting traffic control devices for the selected roadway situation
  • Develop a structured method for selecting traffic control devices that researchers can use to evaluate the selection of traffic control devices for roadway situations not investigated in this study.

Task Descriptions:
Task 1 – Conduct Literature Review
The purpose of this task is to review the state of practice in selecting traffic control devices, to review applications of decision theory to traffic control devices, and to review relevant decision theory modeling techniques. The findings from this investigation will guide choices made in future tasks. Subject areas of interest in the literature review are – present guidance on selecting traffic control devices, descriptive decision theory models, normative decision theory models, data requirements for decision theory models, and statistical methods for creating decision theory models.

Task 2 – Define Investigation Scope
The purpose of this task is to limit the scope of this research to a specific roadway situation, such as traffic control decisions on urban arterials. By focusing on a specific roadway situation, the research will be able to focus on depth, which will advance the body of knowledge on selecting traffic control devices. Important considerations in the selection of a specific roadway situation are – availability of historical data, availability of area experts, and the number of traffic control devices available for use on the selected highway element.

Task 3 – Collect Historical Data
The purpose of this task is to collect documented transportation-engineering decisions on traffic control devices for the roadway situation selected in Task 2. This research will use these data to develop a survey of professionals (Task 5) and to develop descriptive decision theory models (Task 6). This task will focus on collecting engineering plans and specifications that document which traffic control devices engineers selected for use and engineering justification for their use. As part of this investigation, researchers will interview engineers responsible for the plans and specifications, obtaining additional information needed for decision theory modeling; important additional information is the objective achieved by selecting a series of traffic control devices.

Task 4 – Define Objectives for Traffic Control Device Selection
The purpose of this task is to use the data collected in Task 3, and input from area experts, to define a series of objectives for traffic control devices use in the roadway situation selected in Task 2. Possible objectives are – reduce crashes, increase travel time, decrease wrong-way movements, increase driver yielding, etc. The roadway situation selected will define appropriate objectives for selecting traffic control devices. Researchers will use the results of this task to develop a survey for professionals in task 5 and to develop normative decision theory models for the selected highway element in task 7.

Task 5 – Develop and Conduct Survey of Professionals
The purpose of this task is to develop and conduct a survey to evaluate professionals’ ability to select traffic control devices for the roadway situation selected in Task 2. This survey will focus on common design considerations found within the historical data. The purpose of this survey is to compare decisions of area experts to those with less experience. The results of this investigation provide insight into what design considerations this research should highlight in the nominal decision theory models, developed in Task 7. Additionally, this survey assists the profession by indicating areas where experts make decisions differently from professionals with less experience, which indicates the misconceptions professionals with less experience might have.

Task 6 – Develop Descriptive Decision Theory Models
The purpose of this task is to create descriptive decision theory models for the roadway situation selected in Task 2. Descriptive decision theory models indicate how decision makers currently make decisions or use to make decisions. These models will connect the decision to use a traffic control device to the objective sought by the agency. To say it another way, these models will determine which traffic control device increases the probability of an agency meeting its objectives while also avoiding negative consequences. For example, these models can answer can help agencies determine which traffic control device is most likely to increase driver yielding at pedestrian cross walks while also minimizing driver delay and fitting within the agency’s budget.

Within this task, researchers will develop decision theory models based upon historical data collected in Task 3 and models based upon the results of the survey conducted as part of Task 5. The models based upon historical data represent the state of practice in selecting traffic control devices ignoring the experience of the engineer selecting the devices. The models based upon the results of the survey conducted as part of Task 5 represents the way professionals are currently making traffic control device decisions; this represents where the built environment is headed in the near future. An important consideration in this task is which decision theory modeling techniques to use. Within this study, these models will provide insight into the differences among decision by professionals with limited experience, professionals with extensive experience, and the current built environment. Additionally, with these models as a basis, researchers will develop nominal decision theory models in task 7.

Task 7 – Develop Nominal Decision Theory Models
The purpose of this task is to develop statistically valid nominal decision theory models. Nominal decision theory models are models that describe how decision makers should make decisions. Within this task, researchers will develop nominal decision theory models based upon logic structures within decision theory, based upon the results of the survey conducted as part of Task 5, and within the limits specified within the MUTCD. A key component of these models will be a linking of agency goals and objectives to the selection of traffic control devices. For example, these models will indicate a traffic control device that maximizes driver yielding to pedestrians while limiting the increase in vehicle delay resulting from such yielding. Additionally, these models will allow agencies to look at tradeoffs related to these objectives and the cost of the traffic control device. For example, these models will provide agencies a cost per percentage increase in driver yielding behavior.

Key considerations in this task is the logic structure used to make traffic control decisions, selecting appropriate inputs and outputs from the model, and finding an appropriate way to incorporate cost into traffic control decisions. The results of this task, and task 8, will provide a fundamental methodology for selecting traffic control devices. The results of this task would represent a chapter within a well-structured document on selecting traffic control devices.

Task 8 – Develop Structured Method for Selecting Traffic Control Devices
The purpose of this task is to lay out a structured method for selecting traffic control devices using the nominal decision theory models developed in task 8 as an example. A key component of this task will be recommendations related to highway design elements that should be looked at in future research and a layout of how researchers should go about developing models for the recommended highway elements. These recommendations will guide future research in a direction making it possible to create a well-structured document to aid transportation professionals in the selection of traffic control devices.

While Task 7 represents a chapter within a well-structured document on selecting traffic control devices. The results of this task represent a manual defining the methods for developing additional chapters within the well-structured document. For example, the results of this task would be the equivalent of documents explaining how to develop safety performance functions and crash modification factors for the Highway Safety Manual.

Task 9 – Document Findings
The purpose of this task is to document the findings of this research in a final report, which researcher will provide to SWUTC at the conclusion of the project.


Implementation of Research Outcomes:
It is beyond the scope of the Manual on Uniform Traffic Control Devices (MUTCD) to provide the breadth of knowledge necessary for evaluating traffic control devices (TCDs) as part of the larger transportation system.  With this research, investigators used existing theory and a survey of transportation professionals to develop a new decision support tool for use in selecting TCDs in a consistent manner.  Such a process will have value to the profession in the long term, but will take time to implement as it represents a different way of doing things. Further use is also dependent upon acquiring and refining the necessary data to support broad-based implementation.

Impacts/Benefits of Implementation:
This research developed a process for improved performance-measure based selection of traffic control devices  leading to enhance public support for and compliance with, traffic control devices in the field.  Using the procedures developed by this research also provides for better economic based decisions.

Web Links:
Final Technical Report