Elements of Business Rules and Decision Support Systems within Integrated Corridor Management: Understanding the Intersection of These Three Components

CHAPTER 1. INTRODUCTION

The purpose of this guidance document is to provide a fundamental explanation of the "decision support system" (DSS) concept and the relationship of DSS systems and business rules to the integrated corridor management (ICM) community. The document will provide examples and background within and beyond transportation, as well as a clearly organized discussion concerning the elements of business rules and DSS within ICM corridors.

An ICM system is a multimodal transportation management system designed to enhance the accessibility and efficiency of traveler infrastructure (routes and modes, but also information) by directing travelers in the short term to alternative routes and/or modes in a transportation corridor. A corridor is herein defined as a unique travel "wedge" or linear corridor that is anchored by a major highway and supported by a network of alternative routes and modes (e.g., commuter bus, rail, bike, transit, etc.). Inter-regional commute trips and intra-regional "bustle" trips that originate near to (and therefore, are considered "anchored by") the major spine highway, are said to be germane to the "travel shed" of that corridor boundary; that is, they would not go out of their way to use another spine highway to travel "downtown" and back.

A DSS is an information system that supports organizational decision making and, in the case of an ICM, assists in multimodal transportation operation decision making in real-time (Lukasik et al, 2011). To assert a multimodal transportation management structure and control the operation of all participating agencies, "business rules" are required. Business rules are predefined and agreed-upon organizational and inter-agency permissions, constraints, or criteria that bind the participating agencies and affect the DSS solutions.

This document is not a step-by-step guide to developing business rules, criteria, or algorithms (or code) to be implemented within a DSS. Extensive search, contacts, canvassing, and discussions have not turned up detailed specifications of business rules as part of DSS within an ICM (and transportation) domain. Consequently, general principles are outlined for the reader. The document is also not a detailed overview of ICM or DSS history, development, and current state. There are numerous other documents whose focus is on these topics (and several of them are cited in the appropriate areas below). This guidance document attempts to provide an overview of the intersection of these three components, which is an often-overlooked area, and we hope the information provided below is helpful.

DEVELOPMENT OF GUIDANCE DOCUMENT

This document is divided into five chapters, each focused on the business rules as related to DSS for managing integrated corridors.

• Chapter 1. Introduction. This chapter provides the overall purpose, objectives, guidance document overview, intended audience, and a brief background of ICM, DSS, and business rules.
• Chapter 2. Guidance for business rules and DSS within the context of ICM. This chapter describes how DSS was developed in the two U.S. Department of Transportation (DOT) demonstration sites, Dallas and San Diego, as well as how users interact, communicate, and coordinate with colleagues and agencies.
• Chapter 3. Case studies and lessons learned in ICM with emphasis on DSS and relevant business rules. This chapter describes operators' strengths, limitations, biases, and lessons learned from site implementation.
• Chapter 4. Examples of operational decisions with and without business rules constraining options.
• Appendixes. Listing the phases of initiative ICM and resources for the development of concept of operation, and interagency agreement approaches and examples.

INTENDED AUDIENCE

This guidance document presents information on DSS and ICM with a focus on proper business rules. It is intended to be used by practitioners, managers, and designers who have varied levels of decisionmaking authority, ranging from leading initiatives to operations supervisors to shift workers; therefore, they will face a variety of challenges, including overcoming interagency and inter-jurisdictional communication issues, or being fully familiar with decision processes in the context of transportation management. These staff members will be from varied backgrounds, including those who are new to transportation.

BACKGROUND

This section provides a brief overview of ICM, DSS, and business rules. It is meant only to give a high-level overview and introduction with references to more detailed documents about ICM and DSS for the interested reader to pursue. This background is necessary to provide the foundation for discussion of these three topics and the role of business rules.

Integrated Corridor Management

The ICM concept has been defined in the San Diego Concept of Operations report as:

...the operational coordination of multiple transportation networks and cross- network connections comprising a corridor and the coordination of institutions responsible for corridor mobility. ICM programs provide better information, coordination of network junctions, proactive management of capacity and demand, advanced technologies and systems, and improved institutional arrangements. ICMS is a "system of systems," i.e., a transportation management system (TMS) that connects the individual network-based TMS, provides decision support, and enables joint operations according to a set of operational procedures agreed to by the network owners. ICMS facilitates ICM programs to meet corridor needs and realize the ICM vision. (San Diego Pioneer Site Team (2008), page 1-1).

ICM has been a concept for over 10 years but was not fully vetted and implemented until the U.S. DOT "ICM Initiative" was undertaken to identify several "pioneer sites" and ultimately award full demonstration rights to Dallas, TX, and San Diego, CA.

Managing an integrated corridor is possible if agencies take advantage of telecommunications, internet, software, and information technology (IT) capabilities as well as, and perhaps most importantly, the increasing availability of archived and real-time "big data." Together, these resources allow agencies' traffic management systems to archive and share data in real-time (or compare it to historical precedence) to enhance their ability to monitor, react to, and even predict traffic conditions to inform motorists and system operators of any severe delays and the options available for either avoiding problem areas or mitigating the cause.

Use of the term "integrated" implies the need for interagency and inter-jurisdictional cooperation and the ability for proactive traffic management systems to share information with corridor agencies and service providers. ICM sites are complex environments characterized by constant upgrades of technology along with the concomitant demands (e.g., monitoring ongoing traffic operations, responding to emergencies and unpredictable weather, communicating with partner entities and the public) on operators, transportation planners, managers, and designers. Such an environment also calls for constant decisions to be made at the individual and organizational level along with proper protocols for communication between entities.

Decision Support Systems

DSS are primarily computer-based information systems that originally were developed in the management and organizational decisionmaking arenas. They are typically used to sort, rank, or choose alternatives. These DSS operated at the operational and planning levels and were used by upper management to plan strategic and long-range approaches. For more historical information and a detailed bibliography on DSS, which is outside of the current scope of this report, see Marakas (1999); Turban, Aronson, and Liang (2008); and Burstein and Holsapple (2008). A DSS consists of three major components:

• Expert Rules.
• Prediction (Model).
• Evaluation.

Because only Dallas and San Diego were fully implemented at the time of this report, this guidance document focuses on those two regional ICM sites as examples.

Business Rules

A key component of successful transportation systems management is that all relevant agencies both leverage their operating powers as well as achieve administrative efficiencies and cost savings. The communication and coordination of information sharing capabilities can promote interoperability. Too often, the work of the agencies unnecessarily overlaps due to the lack of an interagency cooperation agreement, leading to redundancies. This inefficiency due to lack of interagency cooperation can result in higher administrative and congestion costs.

This guidance document outlines elements of business rules and DSS used in ICM to ensure efficient operations and interagency cooperation.

When discussing DSS, business rules, ICM, and how these concepts tie together, it can often be confusing. One analogy that may help describe this complex relationship hails from chess. In a game of chess, there are several components: the chess board, pieces, rules, and strategies for using those pieces within the rules (see Figure 1 for an example chess board).

Similarly, we can map these components onto the DSS and business rules within the ICM space:

• The ICM corridor can be thought of as the chess board where the pieces interact, decisions are made, and overall context is provided within the geographical constraint of the board, or in the context of ICM, the "corridor".
• The managers, operators, agencies, and organizations involved in the ICM corridor can be thought of as the chess pieces as they are the "moving parts" on the game board/corridor.
• Business rules are the pre-agreed game rules by which these individuals and agencies (pieces) interact. In the chess analogy, a bishop can move diagonally any number of unimpeded spaces, a pawn can move one space forward, etc.
• The DSS contains the strategies to beneficially utilize these rules. In other words, knowing the rules of how the entities interact is not enough, just like knowing the rules that govern pieces on a chess board does not make someone a chess champion. Combining these rules within the chess board context in a most beneficial way to succeed is one way of defining strategy.

Another example could be a hospital operating room. DSS have been in use in hospital operating rooms for many years, evolving from "management systems" over recent decades. There are many task-related rules in place merely to record a patient's vital signs. Various professionals follow protocols to take and monitor the patient's temperature, blood pressure, oxygen level, heart rate, respiration, etc. In case of failure of one or more of these functions, the surgeon plans decisions relating to the operating procedure and makes the necessary adjustments.

In the transportation management arena, we are advocating that traffic management systems have a decision support system that automates much of the necessary decisionmaking based on conditions. But, within this context, it is crucial to have these decisions align and be consistent with agreed-upon policies, procedures, control plans, interagency agreements, and protocols of the lead and cooperating agencies.

Integrated Corridor Management, Relevant Components, and Recent Implementations

ICM is an integrated network of freeway, arterial, transit, and parking facilities, among other transportation networks. Real-time situational awareness on the entire network and all constituent parts is required to make proper decisions. To meet this goal, interagency cooperation is needed.

The existing agencies do very well with day-to-day operations. What causes problems with the corridor operations are atypical non-recurring events, such as incidents, and even atypical recurring conditions. Multi-jurisdictional responses were not being attempted before ICM (Spiller, et al. 2014, page E-4).

ICM gives agencies the opportunity to operate together by sharing information pertaining to incidents, construction, special events, transit, parking, and traffic flow data internally and externally. The shared information is used to improve situational awareness, corridor performance, and real-time alternatives for transportation users, enabling them to manage their trip plans (Miller et al., 2015). Figure 2 lists characteristics of a transportation network where an ICM could be effective:

However, the Federal Highway Administration's (FHWA) Integrated Corridor Management Implementation Guide (Gonzalez et al., 2012) suggests the leaders of each agency answer these questions before they engage in an ICM project:

• Why the ICM is needed?
• How will ICM address the problem?
• How will ICM benefit the users as well as the partners?

As noted above, to date, the San Diego I-15 and Dallas US 75 corridors are the only two fully operational U.S. DOT-funded ICM demonstration sites with DSS systems. However, elements of ICM concepts have been developed and tested in several other locations:

• New York/New Jersey/Pennsylvania – since the mid-1980s, this region has cooperatively championed the TRANSCOM regional authority, which has grown to include a coalition of no less than 16 regional transportation agencies in New York, New Jersey, Pennsylvania, and Connecticut. TRANSCOM was the first de facto ICM-styled, coordinated, multi-regional coalition.
• Minneapolis – the I-394 corridor was a U.S. DOT Pioneer Site for Stages 1 and 2 of the Federal ICM program. The corridor is bounded by Highway 55 on the north and Highway 7 on the south, and is ripe with advanced and robust infrastructure, highway traveler information systems, and real-time data.
• Phoenix – The Maricopa Association of Governments champions the ICM strategic plan for I-10, Loop 101, and potentially the I-10/I-17 "spine" connector. "AZTech" is a strong partner in developing the management partnership of many agencies, cities, and local authorities (Spiller,et al. 2014).

ICM improves communication by supporting coordination among agencies such that they share incident, construction, and special event data and information as well as making changes to benefit the corridor's operation through a data interfacing platform (i.e., a common operational and visualization platform) that can be integrated into the agency's existing software. A DSS, as a major component of ICM, supports the operation of agencies participating in the integrated corridor system by potentially recommending traffic operations strategies, traffic control plans, and response plans and actions and by disseminating information based on prevailing conditions within the transportation corridor.

As ICM builds upon regional management, it requires the expansion of existing agreements among institutions to improve corridor operation. Operations are coordinated through the ICM network, where participating agencies share data and information while making changes that benefit overall corridor operations and performance. An example of this would be operations personnel adjusting traffic signals and ramp meters to take advantage of capacity on other facilities; e.g., funneling travelers to parallel roads, or high-occupancy toll (HOT) lanes or bus rapid transit, as needed. A properly deployed DSS can also compare real-time data against historical data to predict short-term corridor performance problems to recommend (and implement) responses that select the most beneficial combination of ICM strategies before the actual predicted breakdown, as in, "why wait for the actual problem?"

Business rules are rules or agreements that define or constrain some aspect of operations, decisionmaking, and strategy, among several organizations. A DSS that incorporates business rules will incorporate agreements with relevant entities when making these recommendations.

To expand the agreement to facilitate multi-agency collaboration, the first step is identifying the ICM partners. ICM partners can be listed in multiple major categories as shown in Figure 3. Due to multi-agency collaboration, ICM has a wide range of applications which require agencies' coordination, cooperation, and communication. For example, the following list notes areas of deployment in the Dallas ICM:

• Responsive traffic signal system.
• Arterial street monitoring system.
• Agency, Bluetooth, and third party data.
• Transit signal priority.
• Transit parking management.
• Real-Time transit vehicle information.
• Freeway and high-occupancy vehicle (HOV) systems management.
• Weather system reporting.
• Decision Support System.
• 511.

By its nature, ICM consists of the corridor sub network as well as the corridor itself. For instance, the Dallas ICM demonstration site on US-75 is illustrated in Figure 4, where the black line represents the corridor influence area. "The primary criteria were to select a congested corridor where there are both transit and other alternative routes available for diversion in the case of an incident. US-75 is fully built out, with no ability for new capacity additions" (Spiller, et al. 2014, page E-3). The corridor network includes a freeway with continuous frontage roads, managed HOV lanes, the Dallas North Tollway, 167 miles of arterials, a bus network, a light rail network, 900 signals, multiple traffic management centers (TMC), and a regional advanced traveler information system.

The San Diego demonstration site on I-15 is illustrated in Figure 5.

The I-15 corridor is a 21-mile freeway segment in San Diego County. This corridor is a regionally significant segment of I-15 from State Route (SR) 52 in the City of San Diego to SR 78 in the City of Escondido. The corridor also includes the portion of SR 163 from SR 52 to I-15 in the City of San Diego. The I-15 corridor is presently an eight- to ten-lane freeway within the corridor boundaries, with additional auxiliary lanes throughout the corridor. Near the southern section and within the median of I-15 from SR 56 is a two-lane reversible, high occupancy toll (HOT) facility. Known locally as the I-15 Express Lanes, this eight-mile, barrier-separated facility operates in the southbound direction during the a.m. peak period, northbound during the p.m. peak period, and all day (northbound) during the weekend. The existing I-15 corridor can be split into three distinct segments: north, middle and south as shown in Figure 5 (San Diego Pioneer Site Team, 2008. Page 3-1).

The role of DSS in the general ICM context is to:

• Receive data from an information exchange system (e.g., Dallas SmartNET/SmartFusion).
• Evaluate various response plan options.
• Provide recommended plan to ICM coordinator, partner agencies, and information exchange system, which then make the ultimate decision and carry out appropriate strategies.

Following confirmation of candidacy, the first step to initiate an ICM is to develop the concept of operation (ConOps). Appendix B provides a list of references for how to develop a ConOps and examples. "The ConOps does not delve into technology or detailed requirements of the ICMS, but it does address the operational scenarios and objectives, information needs, and overall functionality. The ConOps must also address the "institutional" environment in which integrated corridor management must be deployed, operated, and maintained" (San Diego Pioneer Site Team (2008), Page 1-3). To develop the concept of ICM operations, a sequential order process is suggested by the San Diego I-15 ICM team. The components of this procedure are as follows:

1. System vision.
2. System goals and objectives.
3. Operational concept.
4. Approaches and strategies.
5. User needs.
6. Implementation issues.
7. Institutional framework.
8. Operational scenarios.

Figure 6 illustrates the Dallas ICM overview. The development of the system requirements incorporated the user needs that had been identified during the concept of operations development. Any operational evaluations that are conducted to assess the potential benefits of the system would then be based on strategies and approaches identified in the original concept of operations (Dion and Skabardonis, 2015). More information about the different phases of the ICM initiative can be found in Appendix A.

Overview of Decisionmaking and Decision Support System Design

Why do we need DSS? Humans are prone to numerous biases that affect decisionmaking and can benefit from a support structure.

Factors Affecting Decisionmaking

Decisionmaking is a complicated process with a myriad potential influences that should be considered. These influences can affect the decisions of individual operators, including those who are in leadership positions and drive overall organizational strategies. They also are part of the motivation for utilizing DSS (i.e., avoiding many of these biases or influences). Torma-Krajewski et al. (2010) and Kelly (1999) identified a variety of factors that can impact an operator's decisionmaking, many of which can lead to sub-optimal outcomes:

• Fatigue (including sleep deprivation).
• Stress, often related to perceived or real-time pressure.
• Erratic eating habits (leading to high or low blood sugar).
• Caffeine.
• Lack of information, conflicting information, and uncertainty.

Stress can negatively influence decisionmaking and should be mitigated when possible. Robinson, et al. (2017) found that decision makers under stress can exhibit a variety of behaviors. For example, they may:

• Seek out certainty, be less tolerant of ambiguity, and look for fast choices.
• Experience greater conflict in social interactions.
• Develop "tunnel vision" (narrowed perception due to sensory overload).
• Experience distorted perception.
• Have a decreased ability to handle complex or difficult tasks.
• Focus on short-term survival goals, sometimes at the expense of long-term benefits.
• Choose riskier alternatives.

For these human factors, a DSS offers an objective – not subjective – analysis to a situation and promotes one or more equally objective solutions.

Understanding Common Decisionmaking Biases

Understanding Framing. A very common decisionmaking bias involves people reacting differently to information depending on the phrasing, context, or "framing." (Tversky et al., 1981) This bias can have profound impacts on a transportation system, especially when switching or transitioning operations during a major incident or emergency if the personnel do not adjust their thinking appropriately. A strategy to mitigate this bias is to change labels, colors, or codes to indicate clearly the context has changed (e.g., from normal operations to emergency operations). A DSS can provide guidance on proper label switching and information dissemination depending on the "mode" of operation. In addition, one should be aware of how information is presented and whether it may be framed in a negative or positive way.

Understanding Confirmation Bias. People often favor or seek out information that confirms a prior hypothesis or belief. This is known as "confirmation bias" (Wason, 1968). This bias can affect operations when operators and managers focus more on data that support an initial approach or only listen to opinions that support their plans. Thus, operators may not seek out alternate explanations and inadvertently ignore other useful information. Instead, the correct decisionmaking process or a properly deployed DSS would be to sample the full range of both negative and positive possibilities rather than just the positive ones.

Understanding Anchoring Bias. Individuals have the tendency to rely on the first piece or limited pieces of information when planning or forming an estimate. This is known as "anchoring" (Ariely, 2008, Tversky et al.,1973). This bias often manifests itself in operational situations where the first incoming field reports will drive estimates or the more salient images will affect planning. To mitigate this bias, one should be careful about weighting early or limited information and generate alternative or counter factual options. Another option is to constantly refine estimates as data becomes more reliable over time (Robinson et al., 2017). A DSS can handle the updating of information and properly weigh it to compensate for this bias.

Decision Support Systems—High-level History and Relevant Approaches on Implementation and Management

Decisions are not made in a vacuum, a concept which is often lost in the design and implementation of DSS. This is particularly relevant in the case of ICM and the overlay of business rules for operations, where various entities can affect one another during the process (e.g., local roads being congested during rush hour due to an incident on the highway).

How Do Decision Support Systems Help Overcome Bias?

DSS can be defined traditionally as computer-based information systems that support business or organizational activities and can be fully computerized, human powered, or a combination of both (Robinson, 2016). DSS can also occur in a range of technology levels, from mechanical to digital. Figure 7 illustrates the evolution and examples of DSS, which began with decision support tools that facilitated decisionmaking tasks (before being developed to larger, more complex decision support systems).

Within a transportation context, one can think of a traffic simulation model as an example of a tool that supports data analysis. It is not a system, per se, but rather produces information that supports the process of making decisions. A range of decision support tools are deployed within a transportation context and will continue to be used to manage and control traffic as well as coordinate amongst staff members and outside stakeholders.

Markalas (1999) defined three components that are often considered to be integral to a DSS:

1. The data/knowledge base.
2. The model (criteria and decision context).
3. The interface.

It seems most implementations neglect the context (which in this case can be the business rules governing agency interactions). The current effort focuses on this very important component of DSS.

The word "integrated" can also be used to describe an effective DSS implementation; it represents the unification of all parts within the context of decisionmaking. A DSS is particularly useful for unstructured or semi-structured problems or a litany of biases and problems that occur every day. DSS have been successfully developed for a variety of fields, including clinical decision support and medical diagnosis, finance, business management, agricultural contexts, law enforcement and the military.

DSS in Broader Transportation Areas. As the transportation network and control centers have become increasingly complex, a DSS can play a greater role in optimizing the efficient and safe movement of people and products. The following passage highlights the burgeoning amount of new data streams and expanding complexity facing operations personnel that demands various decision support systems:

In the world of transportation systems operations, emerging infrastructure- based sensor technologies and in-vehicle technologies are providing new data streams to support transportation operations decisionmaking. Increasingly complex and capable system control technologies and traveler systems present transportation managers with a broader range of potential actions to impact system performance. In many cases, this includes a new capability to act with increasing precision with a shorter response time. At the same time, there is an emerging recognition that in complex systems individual mode control decisions cannot be made independently. Any single modal decision may impact broader system performance positively or negatively. This leads to the consideration of how concurrent decisions may be made across modal, facility, institutional and jurisdictional boundaries to optimize performance across the entire multimodal transportation system (Lukasik et al. (2011), page 3-1).

Most existing DSS identified were associated with freeway management systems – from a "lessons learned" perspective, these deployments can be extrapolated to future multimodal systems (Lukasik et al., 2011, page 3-4).

The application of DSS in transportation as stated by Lukasik et al. (2011) has variety of developments in real-time traffic management but not limited to:

• Accident response strategy assessments.
• Online travel information systems.
• Predictive travel time calculations.
• Dynamic route guidance.
• Adaptive ramp metering using predictive traffic congestion algorithms.
• Intelligence-based Transit DSS.
• Dynamic emergency vehicle routing.
• Emissions management.
• Urban and interurban congestion management.
• Security threat mitigation and large-scale evacuation management.

DSS technology and methodologies can be categorized into five major groups (Power, 2001 2003; Lukasik, 2011). Figure 8 outlines the DSS methodologies. Discussing these technologies/ methodologies is out of the scope of this report, but further information can be found in Power (2001, 2003).

The following list with transportation ICM appropriate examples was based on Lukasik (2011):

• Table-based DSS (e.g., Toronto Computerized Optimization Model for Predicting and Analyzing Support Scenarios (COMPASS), Kansas City Scout, Georgia DOT NaviGAtor) are data tables or spreadsheets with predefined response plan recommendations and require little or no processing, modeling or analysis. Some may include basic logic to analyze data in the tables, while others are purely lookup tables.
• Knowledge-driven examples include:
  • Expert Systems (e.g., the Caltrans Advanced Transportation Management System (ATMS), St. Louis Gateway Guide). Response plans are based on a set of pre-defined rules and the DSS requires an expert system engine.
    • Custom Rules-based Systems (e.g., Oregon DOT Transport, Pace Transit Operations Decision Support Systems ). The Custom Rules DSS uses specific rules to determine response plans, making it similar to the Expert System DSS. The main difference between the two is that rules are custom built rather than having an expert system engine.
  • Event Scenario Matrices (e.g., Lake County Passage, Michigan ATMS, New Jersey ATMS) identify events on the roadway using a plane coordinate system, and users can respond to the events using the predefined ITS field devices along the roadway in the area.
• Model-Driven DSS incorporate on-line simulation tool integration (e.g., Singapore – Green LInk DEtermination (GLIDE) Traffic Control System, Madrid, Beijing, Milan).
• Data Driven DSS are a form of support system that focuses on utilizing internal and sometimes external data to aid in the decisionmaking process. Sometimes this comes in the form of a data warehouse, e.g. a database designed to store data in such a way as to allow for its querying and analysis by users.
• There are also hybrids of the "Model Driven" and "Data Driven" models.

The "Five Rights." Lessons have been learned from many of these fields, and DSS has worked across several domains within them. For example, Figure 9 illustrates a brief list of "five rights" to keep in mind when implementing a DSS design (U.S. Department of Health and Human Services).

Proper Design and Use of Decision Aids. Decision aids (such as DSS) can help with data processing and improve overall decision quality. This can offset some of the typical biases that arise when making decisions.

Robinson et al. (2017) found that decision aids have a variety of advantages, including:

• Minimizing the influence of biases (such as confirmation bias) on decisions.
• Forcing structure and consideration of alternatives as well as quantitative weighing of options.

However, the same study also found several disadvantages that one should keep in mind when employing decision aids, including:

• Discounting intuition and experience.
• Lack of use regarding inaccurate recommendations.
• Lack of adaptation in unusual circumstances.
• Increased decision time, and difficult to evaluate.

Despite these potential pitfalls, properly designed decision aids can be valuable assets to an operator in the TMC.

Business Rules: Context and Constraints for Implementing Decision Support Systems Interagency Agreements

A key component of successful transportation systems management is that all the agencies involved leverage their operating powers, achieve administrative efficiencies, and function cost effectively. Communication and coordination can promote interoperability, which supports these objectives. Too often, however, agencies conduct overlapping activities due to the lack of an interagency cooperation agreement. Such inefficiency can result in higher administrative and congestion costs.

This guidance document outlines elements of business rules and DSS used in managing integrated corridors to ensure efficient operations and interagency cooperation. By establishing a framework for sharing information and data and for collaborating, the duplication of data collection and extra delays caused by congestion can be reduced. Gordon (2011) found that significant value can be obtained with:

• The right leadership.
• Strong strategic planning.
• Evaluation of the current case.
• Solid contract management.

To achieve organizational change across multi-agency functions requires a holistic strategic plan. Strategic planning to develop business rules commonly includes several major steps:

• Determine the need for ICM.
• Determine the lead agency and point of contact.
• Elect a committed board to assist with planning and design.
• Define the vision.
• Define resources and categorize them by duration (short and long term) and type (service, or support fund).
• Define stakeholders with an opportunity for public-private partnering.
• Plan outreach activities and follow up.
• Define capabilities of each agency.
• Evaluate the existing plans, rules, and services.
• Conduct detailed planning for all possible scenarios: daily operations, freeway incident, arterial incident, transit incident, special event, and major disaster.
• Determine where there is a need for interagency communication and operation improvement and prioritize them (consider all possible scenarios).
• Identify the best way to create a collaboration where there is a need.
• Decide based on the feedback obtained from the board of committee and stakeholders.
• Allocate the resources.
• Gain approval from each agency's authority.

Consideration for ICM in a corridor (i.e., as a congestion or incident-driven mitigation) is typically led by a regional agency, such as a State DOT, a metropolitan planning organization (MPO) or a transit authority, that has responsibility for overseeing a transportation plan in the subject corridor. An early confirmation that said corridor is even an ICM candidate corridor can be made by viewing the U.S. DOT report ICM Candidate Corridor: Implementation Guide and Lessons (Gonzalez et al., 2012).

Another early step in initiating an ICM system is to select a stakeholder that will manage the process, such as scheduling meetings to discuss planning, activities, issues, and risks. The U.S. DOT ICM website (http://www.its.dot.gov/research_archives/icms/index.htm) offers many sources to assist stakeholders in developing a managed corridor system. The "ICM Knowledgebase" (http://www.its.dot.gov/research_archives/icms/knowledgebase.htm) is also a good source for presentations, newsletters, and fact sheets on ICM.

All potential stakeholders should be invited early in the process (Gonzalez et al., 2012). Their level of involvement will be up to them; however, it is advisable to keep those who decide not to participate informed as the efforts progress. The success of the program is highly dependent on the level of engagement and interest among agency leadership because ICM-related activities require the commitment of time and resources. It's advisable for each key operating agency to have a champion who understands ICM and the concept of corridor operation and could lead a diverse team, communicate well, and commit time to the effort (Gonzalez et al., 2012).

The planning and development process should include transportation planners, modelers, and operations personnel from various partner agencies. These stakeholders are key to understanding the context. They can provide guidance into the types of performance measures needed as well  as assisting the organization in understanding how best to track and evaluate system performance to ensure it remains focused on the established goals. When working with multiple agencies, it is also important for terminologies and acronyms to be well defined, understood, and agreed upon among stakeholders. Educating stakeholders and maintaining their engagement over time is another leadership task critical for the success of the program. Routine evaluation of the risks, benefits, and implications of each strategy by the leaders is required. One example of risk could be compatibility of the technology and data collection format, which are necessary for agencies to exchange certain data elements. However, any changes on the interface and type of data being shared could cause problems.

In accordance with sharing information and data across the system, any future technology enhancement regarding data collection must be compatible amongst all agencies.

Proficiency of the program over time can be a challenge. Planning efforts should focus on ensuring the program is successful enough to run over the long-term, not just for the short period of time that funding is available. One way to facilitate motivation and resource sharing could be through peer exchange workshops bringing representatives with the knowledge of ICM implementation in other regions to share their experience (Markiewicz et al., 2016).

To improve interagency operational power, agencies need to share information (including the names of vendors, contract terms and conditions, and points of contact for each group) and real-time data. The traffic management center (TMC) needs integrated information from the

State DOT, county sheriff's department, safety/service patrol, closed-circuit television cameras, detection systems, other TMCs, agencies, etc. Based on the pioneer sites' experience, to eliminate confusion, stakeholders' roles and responsibilities should be documented. For formal agreements among agencies, one or more business cases can be defined, each of which should include an executive summary with the key point of agreement followed by a defined scope and potential duplication of resources and staff, potential monetary value, role, and interagency demand and management of resources (Gordon, 2011). Although not mandatory, agencies may consider developing an agreement such as a memorandum of understanding to share information or services. More information about interagency agreement approaches can be found in Appendix C. The Dallas/Fort Worth regional comprehensive intelligent transportation (ITS) agreement is attached to this document as an example (Appendix D).

The expected operation of the ICM corridor under each possible scenario must be defined in detail. An expanded description of each scenario solution should be provided in detail and broken down to the responsibility of each agency (more information is available in the Concept of Operations for the I-15 Corridor in San Diego, California (Page 5-1 Chapter 5 – Operational Scenarios). Dallas and San Diego defined their scenarios in general categories as follows:

• Daily operation or non-event based operation.
• Moderate and major freeway incident.
• Major arterial incident.
• Major transit incident.
• Planned special events.
• Major disaster.

The effect of these capabilities can span the institutional and operational processes; however, agencies must be alert to the impact of duplication when creating a new collaborative system. The new business rules should be approved by the agency's executive leadership, although an agency may choose to require additional approvals (Gordon, 2011). Regular assessment of whether the intended business rules meet the goals for the managed corridor is necessary. In addition, an emphasis on transparency when establishing the rules and strategies will reduce any future confusion when it comes to implementation.

To stimulate realistic operations strategies, the board or committee must constantly work to enhance their understanding of respective capabilities, authorities, and limitations. Examples of limitations could be funding or maintaining trained staff and qualified leadership.