Strategic Issues Facing Transportation, Volume 1: Scenario Planning for Freight Transportation Infrastructure Investment (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

40 Scenario Planning Workshop Design The research team conducted six Future Freight Flows Scenario Planning workshops from November 4, 2010 to June 28, 2011. In order to test the validity and the robustness of both the scenarios and the workshop methodology, each workshop was held in a different location and explored a different set of strategic questions. The workshops were run with a local host organi- zation and were held in the locations listed in Table 6. The six workshops were similar in that each included a diverse set of stakeholders who discussed and debated potential infrastructure investment strategies across different potential future sce- narios through both small and large group activities. While the framework was common, each workshop was unique in that it was designed specifically for that host organization and the respective geographic region. There are nine key design components for running a scenario planning workshop, as follows: • Scope (geographic and planning horizon). • Objective (visioning or evaluating). • Duration (half day, full day, multi-day). • Participants (stakeholder distribution and level). • Strategic Questions (what to have the teams decide or provide input on). • Evaluation Elements (infrastructure segments, corridors, themes, etc.). • Evaluation Mechanism (voting rules and data collection methods). • Scenarios (which future scenarios to employ and what collateral to use). • Debrief (how to present outcomes and to whom). Over the course of the six workshops, the research team tried different designs in order to test the effectiveness of each element. In each workshop, the participants discussed potential investment strategies for their specific regions in question. While each workshop focused on a different region and had a different set of strategies to evaluate, each region’s investment options could be roughly categorized into three common classes: gateways, corridors, and connectors. Gateways are points of entry for freight into the geographic region in scope. These cover water, air, and land (both rail and road), ports, and border crossings. Corridors are the high-volume trunk lines that connect different locations across the region. They consist of highways, rail lines, and waterways. Connectors are the elements of the infrastructure that enable the movement of freight between the production or consumption locations and the corridors. Looking across all of the workshops, we found that the overall priority of investment for future freight flows favored connectors first, then corridors, and finally gateways. There were exceptions, of course. But, in general, the connectors were viewed as being critical to any future freight system, while currently experiencing under-investment. Corridors (mainly highways and railroads) were S e c t i o n 4

Scenario Planning Workshop Design 41 seen as important, but additional funding was not seen as being as critical since they have been the focus of investment for a fairly long time. The gateways were the least favored investment class across all the workshops. Interestingly, though, gateways were seen as being exceptionally important within the Global Marketplace scenarios whenever they were run! Because the Global Marketplace scenario was viewed by the workshop participants both as the most similar to today and the most likely to occur, this finding identifies a potential blind spot in planning where a single future is unconsciously designed for. Confirming the importance of connectors to freight infrastructure, the workshop participants overwhelmingly identified “develop or improve intermodal connections” as the most critical initiative to pursue. This initiative took slightly different forms in each workshop as dictated by the specific freight network of the region in question, but was dominant across all workshops and scenarios. The next closest common initiatives were “develop freight-only corridors” and “standardize regulations to facilitate freight.” Overall, the six workshops demonstrated that government planning agencies could successfully complement their existing methodologies using scenario planning. The final deliverables from this project included a “Scenario Planning Toolkit” that contains all materials needed to run a scenario planning workshop. It is available on the companion DVD package and online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). The remainder of this section is organized as follows. Section 4.1 presents the design of the six workshops, including the duration and objective of each workshop, infrastructure segments evaluated, scenarios used, types of participants, and so forth. Section 4.2 describes the process used for the six workshops. This section presents how workshop participants were engaged in the scenario planning process, questions asked, and tools used for extracting insights from the participants. The results from these workshops are discussed in Section 5. 4.1 Workshop Design This section describes the choices made for various components of the six Future Freight Flows workshops. This section begins by describing the nature of engagement between the MIT research team and the host organization for each workshop (Section 4.1.1). This is followed by a discussion of how the nine design components were selected for each workshop (Section 4.1.2). Finally, we describe the skills required to facilitate a workshop in Section 4.1.3. The gist of this section is contained in more of a workbook format in the Future Freight Flows Workshop Planning Guide (available on the companion DVD package and online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). ID D ate Host O rganization Location DVRPC Nov 4, 2010 Delaware Valley Regional Planning Commission (DVRPC) Philadelphia, PA MNDOT Feb 11, 2011 Minnesota Department of Transportation (MNDOT) and the Metropolitan Council St. Paul, MN WSDOT Mar 9, 2011 Washington State Department of Transportation (WSDOT) Seattle, WA POLB Apr 13, 2011 Port of Long Beach (POLB) Long Beach, CA GDOT May 9, 2011 Georgia Department of Transportation (GDOT) and the Atlanta Regional Council (ARC) Atlanta, GA U . S . DOT Jun 28, 2011 U.S. Department of Transportation (U.S.DOT) Washington, D.C. Table 6. The six future freight flows workshops.

42 Scenario Planning for Freight transportation infrastructure investment 4.1.1 Engagement with the Planning Organization The project team from the MIT Center for Transportation & Logistics (CTL) started the engagement with each host organization about 8 weeks prior to the workshop. Regularly sched- uled phone calls were conducted between MIT and the host throughout the planning period. The workshop dates and locations were usually set in the first call. The later calls were used to make various choices about the design of the workshop. These choices are presented in Section 4.1.2, in rough chronological order. The level of engagement of the host organization in planning the workshop varied across the six workshops and ranged from being intimately involved in all aspects of the planning details to essentially just providing space to run the workshop. As a general observation, we found that the more engaged the host organization was in the design, development, and delivery of the workshop, the more successful it was. The planning agencies at the DVRPC, MNDOT, WSDOT, and U.S.DOT were the most heavily involved hosts. DVRPC took the most effort in the logistics and execution of the workshop— most of the details of the later workshops used the material developed at this first session to include the logo itself. MNDOT, along with the Metropolitan Council and the Volpe Center, focused the workshop design to complement an existing planning project for the twin cities. Together, the three organizations chalked the scope of the project and customized the focus of the workshop on “themes” rather than specific investments. They also took copious notes of the discussions facilitated by the scenario planning session and used the output of the workshop in their ongoing project. WSDOT took much initiative in planning the workshop and selecting the portfolio of attendees. They focused on evaluating entire corridors for freight movement and combined some of the “open-ended” aspects from the MNDOT session. The U.S.DOT sessions tackled the more difficult national problem by introducing components of a freight network (gateways, corridors, and connectors) instead of specific modal infrastructure segments. Also, we utilized interactive electronic polling response tools during this national session. 4.1.2 Workshop Design Components The general framework of each workshop was the same. There was an introduction to the concepts and approach followed by small group immersion and breakout sessions, with a final group debrief and discussion. Within this general structure, however, each workshop was designed differently using nine key components. These are as follows, in roughly chronological order they should be selected: • Scope (geographic and planning horizon). • Objective (visioning or evaluating). • Duration (half day, full day, multi-day). • Participants (stakeholder distribution and level). • Strategic Questions (on what to have the teams decide or provide input). • Evaluation Elements (infrastructure segments, corridors, themes, etc.). • Evaluation Mechanism (voting rules and data collection methods). • Scenarios (which future scenarios to employ and what collateral to use). • Debrief (how to present outcomes and to whom). As mentioned in the opening paragraph, the sections are presented in the chronological order in which they were chosen by the planning organization and research team. The following sections describe each individual design component, discuss how they were selected for each workshop, and provide a recommendation for future workshops.

Scenario Planning Workshop Design 43 4.1.2.1 Scope The selection of the scope should be the first element decided for any workshop. The scope includes the geographic region being considered (state, national, multi-state, etc.) as well as the desired planning horizon (10, 20, 30 years, etc.). The scope was usually agreed upon during the first phone call between the research team and the planning organization—right after the date and location of the workshop was set. The geographic regions differed for each workshop but, with the exception of POLB, each was the area under the jurisdiction of the host organization. POLB used a multi-state geographic region to get at the import/export issues that cross multiple state and jurisdictional lines. The geographic scope for each workshop is listed below. DVRPC: City of Philadelphia and the nine counties (five in Pennsylvania, four in New Jersey) for which the Delaware Valley Regional Planning Commission develops plans. MNDOT: Entire state of Minnesota, with an emphasis on the Saint Paul and Minneapolis metropolitan areas. WSDOT: Entire state of Washington. POLB: All major corridors and ports in the Los Angeles and Southern California area that connect the port with the rest of the U.S. GDOT: Entire state of Georgia. U.S.DOT: The contiguous 48 states of the United States. As discussed later in the report, evaluation elements used within a workshop can fall outside of the geographic region if they affect the area under consideration. For example, for the WSDOT workshop, a Canadian highway was one of the potential investment segments. Similarly, improvements to water ports in neighboring states were considered in the GDOT workshop. The same planning horizon was used for each of the workshops: 20 to 30 years from today. The scenario collateral for all of the workshops used the date of November 2, 2037 for the newscast videos that were shown as part of the scenario immersion process. For future workshops, we recommend that the geographic scope be in line with the jurisdictional control or responsibility of the host organization. This provides more in-depth knowledge for the underlying network and increases the quality of the preparation. 4.1.2.2 Objective (Visioning or Evaluating) Scenario planning workshops are designed to either enable visioning of potential future strategies or facilitate the evaluation, ranking, and selection of a strategy from an existing set of potential choices. Visioning workshops are pure brainstorming exercises that are used to develop new unfettered thoughts or ideas: a “clean sheet of paper” approach. Evaluating workshops, however, require the participants to compare and contrast between a set of alternatives. This forces the participants to make choices and trade-offs, debating the pros and cons of the alternatives with each other. Both workshop objectives have strengths and weaknesses. Visioning workshops are good for areas where innovative thinking is required and no established options are desirable or sufficient. These sessions are great for bringing up new and out of the box ideas, but they rarely drive to a consensus or produce an actionable recommendation. These sessions tend to have very open-ended discus- sions and are more akin to a brainstorming session that generates ideas rather than makes deci- sions. Evaluation workshops are just the opposite. Presenting a closed set of options to a group of participants focuses the participants’ attention and forces them to make decisions. There is a risk of missing potential strategies not initially considered, but providing an “other” category can mitigate this. Collecting, harmonizing, and analyzing evaluation data is much easier than visioning data.

44 Scenario Planning for Freight transportation infrastructure investment Of course, most workshops will combine a mix of these approaches, but, in general, a workshop will lean one way or another. Presenting options in a visioning session can lead to some anchoring of discussion by some participants. The DVRPC and U.S.DOT workshops were pure evaluation exercises: the participants evaluated the chosen infrastructure segments and did not engage in any visioning or open-ended exercises. The U.S.DOT workshop did, however, employ a series of more in-depth evaluation questions that explored funding and other issues beyond selecting investments. These additional evaluation questions served as a starting point for the larger group discussions—and helped to bring out more “visioning” comments in the final debrief session. The MNDOT workshop was predominantly a visioning exercise. The remaining three workshops (WSDOT, POLB, and GDOT) were pre- dominately evaluation workshops with a small amount of visioning in each of the small group breakout sessions. For future workshops, we recommend that if time and schedule permits, the host initially runs an open-ended visioning session with a small, core group of stakeholders. Then, using the sug- gested approaches and strategies from this session as the set of strategies to pick from, they can run a series of evaluation workshops for a larger set of stakeholders. In the evaluation workshops there should still be some open-ended element to capture things that have not been previously considered. As was done in the U.S.DOT workshop, the use of more focused follow-on evalua- tion questions can help generate additional discussion and more open-ended comments. As one host mentioned, the actual selection of the potential investments is not as important as under- standing the thinking and the logic that went into that selection. 4.1.2.3 Duration (Half Day, Full Day, Multi-day) Future Freight Flows scenario planning workshops have been run in durations ranging from half to a full day. The DVRPC was the only half-day workshop that the research team ran. Afterwards, the MIT team realized that a four-hour workshop did not provide sufficient time to understand the scenarios, explore their implications, and evaluate various infrastructure segments and/or generate ideas for strategies in each scenario. All of the remaining five workshops lasted three-quarters of a day or about 6 hours. In all of these workshops, however, we found that the debriefing session of the results was always rushed. We recommend that for future workshops, a second day be included for a detailed debrief of the results from the session to a select group of decision makers. The workshop itself should remain 6 hours—any longer taxes the attention span and attentiveness of the participants. 4.1.2.4 Participants (Stakeholder Distribution and Level) Participation in all six workshops was by invitation only with a priority going to individuals with first-hand knowledge of the region’s freight infrastructure needs. This suggested that government transportation planners in the region, shippers, carriers, as well as community and environmental groups be invited to participate in the workshop. This also suggested that consulting firms and independent consultants should not be invited, unless such a person was deemed to be highly insightful by the host organization. The number of people from each category who agreed to actively participate (which excludes members of the host/planning organization participating as note takers) in each workshop is mentioned in Table 7. Once the workshop date, location, and duration were finalized, MIT and the host organiza- tion started identifying and contacting the potential participants (about 6 to 8 weeks before the workshop). Besides MIT and the planning organization, the candidate participants for the workshops were also sought through recommendations from the members of the research panel.

Scenario Planning Workshop Design 45 The participants from varied backgrounds were invited to attend the workshop. While they were categorized slightly differently for each workshop, the participants fell into three large groups: shippers, carriers, and public sector. Within these groups there were 12 more specific categories. 4.1.2.5 Strategic Questions (What to Have the Teams Decide or Provide Input On) Once they were immersed in their respective future scenarios, the workshop participants were tasked with addressing one or more strategic questions. Table 8 shows the specific questions used for each of the workshops. Participant C ategory DVRPC MNDOT WSDOT POLB GDOT U.S.DOT TOTAL 3PL 8 3 2 5 5 23 Academic 1 2 2 1 6 Association 2 4 1 1 8 Carrier 15 12 8 10 3 9 57 Citizens, environment 4 4 Consultant 5 2 2 3 12 Government—Fed 2 5 1 3 11 Government—Local 1 5 17 1 2 8 52 Government—State 5 8 10 3 5 2 33 Panel 2 4 5 11 Port 6 8 5 9 28 Shipper 8 8 10 1 2 9 13 60 Total : 6 8 53 61 41 36 46 305 Note: 3PL = third-party logistics. Table 7. Number of invitees agreeing to participate (by category). Workshop Objective Strategic Question(s) DVRPC Evaluating “Which infrastructure investment bundles [should] we invest in TODAY to prepare for the scenario in YEAR 2037?” MNDOT Visioning “How should MNDOT prioritize different themes?” “What initiatives within each theme should be pursued today to improve freight infrastructure in 2037?” WSDOT Evaluating “Which freight investment segments will be most critical in 2037?” “What will be the primary (and secondary) freight corridor in 2037?” “What are some initiatives that WSDOT should take to improve this freight c orridor?” POLB Evaluating “Which freight infrastructure segments should be invested in TODAY to be ready for the year 2037?” “What initiatives should be undertaken TODAY to prepare for this scenario?” GDOT Evaluating “Which freight infrastructure segments should be invested in TODAY to be ready for the year 2037?” “What actionable initiatives should be undertaken TODAY to prepare for this scenario?” U . S . DOT Evaluating “Where should we prioritize federal funds NOW given that the future described in your scenario in 2037 is going to occur?” “What level of investment should the federal government take for each type of infrastructure (maintain existing, improve existing, or add new)?” “Where should the policy be made (local or federal) and how should primary funding be provided (private, public, private–public)?” Table 8. Strategic questions for each workshop.

46 Scenario Planning for Freight transportation infrastructure investment The strategic questions for all of the workshops, except MNDOT, were very similar, differing only in the naming of the options (investment bundles, segments, etc.) and some follow-on questions. The MNDOT workshop was a visioning session so its questions were more open ended. Even the prioritization of the freight action bundles (FABs) was more open ended—real trade-offs were hard to make as the FABs consisted of “highway system improvements/congestion management,” “rail, water, and air improvements/congestion management,” “modal balance and intermodal options,” “land-use strategies,” and “policy and regulatory initiatives.” The main point of the strategic questions was to get the participants to think of present-day actions or investments to take given that their assigned future scenario occurs. The participants were typically told to ignore where the funds will come from or the timing or sequencing of the funding. For future workshops, we recommend that the host continue to have the participants vote on the priority of the individual options but that the follow-on questions can be more open ended. Forcing the participants to make hard decisions between the competing alternatives brings up valuable discussion that leads to potential insights on future freight flows. Also, it is important that the host organization capture the decision making behind the specific votes. 4.1.2.6 Evaluation Elements (Infrastructure Segments, Corridors, Themes, etc.) The selection of the elements or segments to evaluate was usually the most discussed component of the workshop design. The elements had to be finalized no later than 2 weeks before the workshop, as they were used in a pre-workshop survey conducted by MIT as part of the research project. Generally, each element consisted of a single-mode contiguous artery used for transporting freight in the defined region. The choice of elements was made by the host organization alone (e.g., WSDOT), primarily by MIT (e.g., POLB, GDOT), or jointly (e.g., DVRPC). As discussed earlier, MNDOT did not have elements per se, but instead created FABs that were more thematic than infrastructure oriented. For the U.S.DOT workshop, the elements chosen were not specific physical elements of the freight infrastructure, but rather mode-independent classes of infra- structure: gateways, corridors, and connectors. In fact, all of the elements used in the five evaluation workshops can be categorized into the three classes used in the U.S.DOT workshop: gateways, corridors, and connectors. Gateways are points of entry for freight into the geographic region in scope. These cover water, air, and land (both rail and road) ports and border crossings. Corridors are the high-volume trunk lines that connect different locations across the region. They consist of highways, rail lines, and waterways. Connectors are the elements of the infrastructure that enable the movement of freight between the production or consumption locations and the corridors. The gateways, corridors, and con- nectors chosen for the evaluation workshops were often, but not always, specified by the mode. The number of segments of each type used in the workshops is listed in Table 9. There was significant discussion with the host organizations as to the number, size, and form of the elements to use. For future workshops, we recommend that between eight and a dozen elements be selected. Using fewer than this does not appear to provide sufficient variety while providing more than a dozen elements to evaluate seems to overwhelm the participants. We recommend that the elements selected be of sufficient size or magnitude to be worth the dis- cussion and be important to multiple stakeholders (improving a specific exit ramp, for example, is too small of an element). Also, the elements should not already be in the current funded investment plan. The challenge is to select elements that are big enough to warrant the discus- sion, but specific enough to avoid generalities (such as, “build more roads”). We like the idea of examining the complete infrastructure in a region and classifying it into the three major catego- ries (corridors, gateways, and connectors) and making sure that each category is represented. If there are multiples of each (such as more than one seaport or airport), they can be collapsed into common group categories without any loss of detail. The distinctions between the multiple

Scenario Planning Workshop Design 47 airports in a category, for example, can be explored during the discussion. Finally, maintenance of the existing infrastructure should not be included as a unique element. We recommend to the host state that it is assumed that maintenance is taken care of separately. 4.1.2.7 Evaluation Mechanism (Voting Rules and Data Collection Methods) There is extensive academic literature on how to design voting or evaluation mechanisms. The rules used within an election, for example, can influence the outcome, encourage or discourage collusion, and shift power to or from minority blocks. Most of the literature deals with the pros and cons of using single-winner versus multiple-winner voting schemes. We focused solely on multiple-winner voting mechanisms since the objective of the exercise is to encourage discussion of all elements—not just a selection of the single most critical element. It is important to remember that the objective is to enable and encourage discussion amongst the various stakeholders—not to elect specific investments. Three different forms of voting were presented as options to the host organizations: single voting, cumulative voting, and ranked voting. Each is explained below with the assumption that there are n elements and m votes per participant. • Single Voting is where each voter places one vote each on up to m different elements. Obviously, m must be less than n. This is essentially a way to pick m “winners” (and thus m-n “losers”) out of n total elements. The advantage of single voting is that it allows preference to be shown for a set of elements without specifically singling out any individual one. Some of the host organizations expressed concern that some government participants would be reluctant to endorse specific investments in the workshop in fear their decision would be considered official policy. The downside of this mechanism is that it does not allow for a voter to distinguish the magnitude of their support for each element. • Cumulative Voting is where each voter places up to m votes on any of the n elements. The value of m can be larger than n. For example, each voter could be given 100 points to allocate as they see fit amongst the n elements. The benefit of this approach is that the degree of support is readily evident in the voting. The downside is that a single voter or minority block of voters can dominate by placing all of their votes on a single choice. Limiting m to be close to or smaller than n can minimize this effect. DVRPC WSDOT POLB GDOT U.S.DOT G at ew ay s Border crossings 2 Water ports 1 1 1 2 3 Airports 1 2 1 1 1 C or ri do rs Highways 3 7 6 4 3† Rail lines 1 3 4 4 Waterways 1 Pipeline 1 C on ne ct or s Local roads / freight connectors 1 1 1 2† Short-line rail 1 1 2 Intermodal facilities 1 1 Number of segments used: 8 16 15 13 12 †The three corridors and the two connectors in the U.S.DOT workshop were specified without any modes. The workshop participants were asked to consider the corridors and connectors to consist of any one or more modes. Table 9. Types of infrastructure segments used in workshops.

48 Scenario Planning for Freight transportation infrastructure investment • Forced Ranking is where each voter places n votes for his or her first choice, n–1 votes for his or her second choice, n–2 votes for his or her third choice, and so on. Each voter is essentially rank ordering the n elements. If all elements are to be ranked, then each voter would need n(n + 1)/2 votes. If only a subset is to be ranked, say, k < n, then each voter would get k(k +1 )/2 votes. This mechanism brings out each voter’s priority, but does not reveal the degree of preference between the elements. This also minimizes the minority block issue. While there was initially a lot of discussion with the host organization on how to vote, we ended up using some form of cumulative voting in each workshop. Professional weight poker chips were used for casting the votes on a large sheet of paper that geographically displays the different ele- ments. Using poker chips forced the participants to get up and physically place their chips around a table, forcing interaction with each other and making the votes more visible. It also reinforced the sense of playing a game and this led to more interesting and open discussions. In all of the work- shops except the U.S.DOT, each voter was given 100 points worth of chips in 25-, 10-, and 5-point increments. Thus, they essentially had 20 votes each. For the U.S.DOT, we restricted the number of votes to 12, the same as the number of elements (12), and each chip counted as one vote. This was a better method since it made counting faster and removed the need to make change. In addition to these “positive” votes, we introduced the concept of “negative” or “veto” votes. Each voter was given three black poker chips—We actually had a skull and crossbones embossed on them for effect!!—that they could use to express their opposition to a specific element. They had to place at least one and up to three negative votes on different elements with only one negative vote per element allowed (single voting). The negative votes were used to force each participant to say no to something. It is a truism in management that having a strategy means saying “no” to something. We wanted to force this aspect in order to get more discussion out of the participants. It was very successful. Negative voting was not used in MNDOT since the FABs were not conducive to this mechanism. For example, we could not envision a voter saying no to “land-use strategies” or “policy and regulatory initiatives.” The FABs were not designed to be traded off. We found that, in general, private-sector participants tended to use more negative votes than public-sector participants. For future workshops, we recommend that a hybrid positive/negative voting mechanism be used with poker chips. For positive voting, a cumulative mechanism should be used with each voter receiving as many chips as there are elements to evaluate (n). For negative voting, a single vote mechanism should be used with each voter receiving three chips of which they have to use at least one. The combination of positive and negative votes worked well in establishing points of discussion during the breakout sessions—which is the sole purpose of the voting mechanism. 4.1.2.8 Scenarios (Which Future Scenarios to Employ and Collateral to Use) Four scenarios were created for the project: Global Marketplace, Naftástique!, One World Order, and Millions of Markets. Millions of Markets was initially called Technology Savior in the first three workshops. The name was changed to remove the anchoring bias of the name to all things technological. Full descriptions of these are not included in this report. Each scenario describes the world assuming a different set of macro sociotechnical and economic factors. The primary driving forces were level of trade (global to regional) and availability of resources (high to low). Whenever possible, we used all four scenarios within a workshop. However, the decision of the number of scenarios to use was based on the number of workshop participants. The optimal size of a breakout section is between 10 and 15 people. Thus, this decision was made as late as about one week before the workshop. Table 10 shows which scenarios were used in each of the six workshops (indicated with an “x”). With the exception of the workshops at the POLB and GDOT, each workshop used all four scenarios. POLB and GDOT each used only three scenarios because to their smaller size. The

Scenario Planning Workshop Design 49 selection of which scenarios to use if the attendance does not justify using all four is arbitrary, but since Global Marketplace is consistently viewed as most like today, we recommend removing this one first. We do recommend, however, that during the debrief session all four scenarios are discussed and presented. Similarly, if the attendance is so large that the breakout sessions exceed 15 people, multiple separate breakout sessions can be run using the same scenario. For future workshops, we recommend that the host organization target a total attendance of 60 people. The number of breakout groups should be determined by dividing the number of confirmed attendees by four. • If this number is between 10 and 15, use all four scenarios. • If this number is less than 10, find the largest number of breakout groups that gives you at least 10 per group: this is the number of scenarios to use. We recommend (not that strongly, though) that the scenarios be used in this priority order: Naftástique!, One World Order, Millions of Market, and Global Marketplace. • If this number is greater than 15, find the smallest number of breakout groups that gives you no more than 15 per group: this is the number of scenario breakout sessions you will need to run. We recommend that the selection of which scenarios to double up on be in the same order as above. The assignment of specific attendees to specific scenarios should be made about a week before the workshop. This was done ahead of the workshop since the brochures were sent out ahead of time. The selection was made such that each scenario had the same number of participants and roughly the same proportion of participants from different categories. When it was not possible to evenly distribute members of one category across all chosen scenarios, similar categories were combined and then the participants from the broader categories were randomly assigned to the selected scenarios. This method of sampling is known as stratified sampling, and is used in order to have a diverse perspective represented in each scenario and to have a mix of perspectives across all selected scenarios. The same collateral should be used within each breakout session regardless of the number of scenarios being run. This should include the respective brochure and the newscast video. 4.1.2.9 Debrief (How to Present Outcomes and to Whom) In each of the scenarios, we held a debrief session with all of the participants present. Additionally, for some of the workshops we held informal post-debrief sessions with the host organization members alone. The specific results will be discussed in Section 5. For future workshops, we recommend a two-tiered debrief approach. First, we recommend that the host organization run a “public” debrief during the workshop. This should follow the same format as was done in the six workshops. The objective is to give the participants an idea of the consensus and differences in investment strategies between the future scenarios. If facilitated correctly, this session is great for getting insights from the attendees. Second, we recommend a more formal and longer “private” debrief be held for just the host organization one or more days later. This session would go into much more depth and the objective DVRPC MNDOT WSDOT POLB GDOT U.S.DOT Global Marketplace X X X X X Millions of Markets X X X X X Naftástique! X X X X X X One World Order X X X X X X Table 10. Scenarios used in future freight flows workshops.

50 Scenario Planning for Freight transportation infrastructure investment would be for the host to wring as much insight as possible out of the results. By holding it a few days later, it allows the team to clean up and spend more time with the results of the sessions. 4.1.3 Workshop Facilitation One of the deliverables for the Future Freight Flows project is a facilitator’s guide that will outline how a scenario planning workshop should be conducted. A team of researchers at the MIT CTL, having the experience of conducting dozens of scenario planning workshops, facilitated all six workshops in this project. For each workshop, the team prepared a facilitator’s guide, which all facilitators followed. These can be found in Appendix E, Exhibits 2 through 7. After the first workshop, the MIT team varied the process followed in the subsequent work- shops based on what had worked well previously and what needed to be improved. In three workshops (DVRPC, MNDOT, and WSDOT), the staff at the host/planning organization or the agency’s collaborators helped the MIT facilitators during the scenario breakout sessions by taking notes and answering any participants’ questions about the freight transportation infrastructure in the region being considered. For example, associates from the Volpe Center worked with the associates from MNDOT and the Metropolitan Council at the MNDOT workshop. Initially, we had planned for the facilitator role to be taken over by the host organization. The MIT team had planned to develop and use a “train the trainers” approach over the course of the six workshops. This was not successful. We found that while the host organizations had domain and local expertise on the freight network, they did not typically possess the required group facilitation skills. We inadvertently discovered that having non-host affiliated facilitators led to a more trusted session. The MIT facilitators, for example, had no vested interests in the outcome of the work- shop. Some hosts mentioned that if members of the local DOT facilitated the session, it would be perceived that they were biasing the discussion to their preferred investment decisions. For this reason, we recommend having non-interested third-party facilitators for future workshops. 4.1.4 Summary This section presented the design of the six scenario planning workshops in the Future Freight Flows project. We pointed out the similarities and differences in the designs of the six workshops along the nine key dimensions: Scope, Objective, Duration, Participants, Strategic Questions, Evaluation Elements, Evaluation Mechanisms, Scenarios, and Debriefs. The next section describes the process that brought these different elements together to facilitate a strategic conversation about investing in the region’s freight infrastructure to be prepared for an unknown and unpredictable future 20 to 30 years from now. 4.2 Future Freight Flows Workshops: Process This section describes the specific scenario planning process used at the six workshops. There are four subsections. The first subsection (4.2.1) presents the resources used in the workshop. This consists of the “human resources,” that is, the roles performed by various actors in the workshop, facilities and equipment, and reading and voting material. Sub sections 4.2.2, 4.2.3, and 4.2.4 present the activities performed before, during, and after the workshop, respectively. 4.2.1 Resources Used in the Workshop This section describes various resources used in the six workshops. This section first describes the roles played by various members of the MIT team and the host organization. This is followed

Scenario Planning Workshop Design 51 by the facilities and equipment used for the workshop, folders prepared for individual participants, scenario videos, and the instruments used for voting. 4.2.1.1 Roles The personnel from MIT and the host organization performed eight roles. The roles are described below. • Host: The host, a high-level executive at the host/planning agency, whose role was to welcome the workshop participants, and inform them of the importance of the workshop for which they were going to invest a day of their time. The following were the hosts at the workshops: – DVRPC: Mr. Barry Seymour, Executive Director of Delaware Valley Regional Planning Commission. – MNDOT: Mr. Bill Gardner, Director of Freight, Rail and Waterways for Minnesota Department of Transportation. – WSDOT: Ms. Paula Hammond, Secretary of the Washington State Department of Transportation. – POLB: Mr. Eric C. Shen, Director of Transportation Planning for the Port of Long Beach, California. – GDOT: None. The Georgia DOT preferred to participate in the session but not publically endorse it. This decision was announced to the MIT team a few weeks prior to the workshop. – U.S.DOT: Mr. John Horsley, Executive Director of the American Association of State Highway and Transportation Officials (AASHTO). • Planning Manager: The planning manager was the one who had been engaged with MIT in designing the workshop for the 8 to 12 weeks leading to the workshop day. In many workshops, the planning manager—Ms. Barbara Ivanov at WSDOT being a great example—took the leadership role of ensuring that the insights collected at this workshop would be brought into the organization’s planning process. In all workshops, except the POLB and GDOT workshops, the planning manager described the infrastructure segments to the workshop participants. • Lead Facilitator: The role of the lead facilitator was to set the stage for a productive scenario planning session. At the beginning of each workshop, the lead facilitator introduced the workshop participants to the philosophy of scenario planning and described how the workshop would be conducted. For all six workshops, Dr. Chris Caplice of MIT played this role. • Scenario Facilitator: The role of the scenario facilitators was to facilitate the discussion within their breakout groups. The main objective of the facilitator was to help the participants in his/her group immerse themselves into the scenario. The facilitator then helped them apply their knowledge and insights to express through a voting mechanism the utility of the can- didate freight segments in their scenario. The scenario facilitator had to manage the dynamic interaction within the group so that individual group members could express their unique insights and then combine the individual insights to bring forth the group’s insights. Five expe- rienced facilitators from MIT—Dr. Chris Caplice, Jim Rice, Dr. Mahender Singh, Dr. Roberto Perez-Franco, and Shardul Phadnis—played this role in the six workshops. • Note Takers: The responsibility of the note takers was to capture the insights being shared by the participants in the breakout session. They were members of the planning organization, and were used in the first three workshops. Each breakout group had one or two listeners. • Associate Facilitator: In three workshops (DVRPC, MNDOT, and WSDOT), the associate facili- tator helped the lead facilitator compile the data generated by individual scenario teams during the breakout sessions. This helped to expedite the cross-scenario analysis so that a fairly thorough analysis could be presented to the workshop participants during the debrief and discussion session. Miguel Sánchez-Valero of MIT played this role. After the WSDOT workshop, the MIT researchers enhanced the spreadsheets used for analyzing the data by the scenario team. This automated much of the analysis, and the role of the associate facilitator was eliminated for the last three workshops.

52 Scenario Planning for Freight transportation infrastructure investment • Cross-Scenario Facilitator: The role of the cross-scenario facilitator was to orchestrate the presentation of results from individual scenarios, engage the participants in comparing the results across scenarios, and to present the cross-scenario analysis. Dr. Caplice of MIT played this role in the first three and the last (U.S.DOT) workshop. Shardul Phadnis of MIT played this role in the POLB and GDOT workshops. • Organization and Reception: Eric Greimann of MIT played this role. Before the workshop, he helped organize the resources needed for the workshop by coordinating with representatives from the planning organization. On the day of the workshop, Eric manned the registration desk, signed in the attendees, and handed out individual folders containing the material used in the workshop. 4.2.1.2 Facilities and Equipment Three types of facilities and equipment were used in all six workshops, as follows: • Conference Room: This room was large enough to seat the entire group of workshop attendees. The room had large projector screens and audio-visual equipment that were used for the PowerPoint presentations. • Breakout Rooms: The breakout sessions were conducted in separate rooms—one for each scenario. Whenever possible, the participants were seated around tables in a U-shape so they faced each other and the facilitator. Each breakout room had audio-visual equipment, which was used for showing the video of the newscast in each scenario. • Audio-Visual (AV) Equipment: Each breakout room had a projector, a screen, and speakers. The AV equipment in the breakout rooms was used to show the video of each scenario (audio needed). The AV equipment in the main conference room was used to show the videos of all four scenarios during lunch and for the presentations before and after the interactive workshop breakouts. 4.2.1.3 Individual Folders Each workshop participant received a folder containing information about the exercise when he or she registered in the morning. The folders were made specific to each scenario. Each folder contained a copy of the day’s agenda, a brief description of freight infrastructure segments (or FABs in MNDOT), maps of the segment (except for the MNDOT and U.S.DOT workshops), and the brochure of the participant’s assigned scenario. 4.2.1.4 Scenario Videos Each breakout group was shown a scenario-specific newscast video. The video first describes the scenario in about 30 seconds, and then shows a newscast from that scenario dated Nov 2, 2037. The video lasts for about six minutes. The reason for using this video is to help the participants immerse themselves in the scenario by showing them a vivid description of the world. After the breakout sessions, all participants saw either a summary video of all of the scenarios, or each of the individual newscast videos of all four scenarios before engaging in a cross-scenario analysis. 4.2.1.5 Voting Instruments Three instruments were used in each breakout session to facilitate voting. • Individual Investment Decision Form: Each participant in the scenario breakout session was given one form to write his/her individual vote. For “evaluation” workshops, the form listed all the infrastructure segments chosen for the workshop, with spaces to write the number of positive (votes) or negative (veto) points assigned by the participant to that segment. For the “visioning” (MNDOT) workshop, this form listed the five FABs considered and a box for writing the points assigned to that FAB. All forms also had a space for the participant to write in the infrastructure segment or FAB not covered in the list provided. The segments on the

Scenario Planning Workshop Design 53 “evaluation” forms were presented together according to their modes. The reason for using these forms was to allow the individuals to write their investment decision based only on their own thoughts and insights before participating in a group discussion and voting process. • Voting Chips: Professional size and weight poker chips were used for the participants to reveal their votes to the entire group. The chips made the votes of the entire group participating in a scenario “visual” and facilitated the discussion about the utility of various segments perceived by the members of the group. For the “evaluation” workshop, the participants were given chips of three different colors, representing three different values (blue for 25, red for 10, and white for 5 points), to represent the positive votes and black chips to represent the veto votes. For the details of the voting mechanism, please refer to the reports for individual workshops in Appendix E. • Group Voting Sheets: A form listing all the infrastructure segments or the FABs was used for the group members to place their voting chips on to represent their individual votes. The scenario facilitator tallied all the votes on each segment or FAB and wrote the total on the form. If any participant changed his/her vote after the group discussed the votes, the changed vote was noted on the form. The final tally of votes from this form was entered into the spreadsheets used by MIT to summarize the group’s vote in each scenario and to compare the votes across scenarios. 4.2.2 Pre-Workshop Activities One week before the workshop, the workshop participants were sent a pre-workshop survey via email. This survey was a part of the research work conducted by MIT during the Future Freight Flows Symposium. The objective of the pre-workshop survey was to capture the work- shop participants’ assessments of the usefulness of investing in various freight infrastructure segments used in the workshop. The survey was conducted in all except the MNDOT work- shop since that particular workshop did not involve any evaluation of infrastructure segments. After completing the survey, each participant was sent reading material for the workshop. This included a web-link to the scenario he or she was assigned to, description of the infrastructure segments used in the workshop, and a one-page description of the workshop. 4.2.3 Workshop-Day Activities With the exception of the first workshop (DVRPC), all other workshops were 6 to 8 hours long. The DVRPC workshop was a half day exercise held from 8 a.m. to 1 p.m. The agendas for all six workshops are presented in Exhibit 1 in Appendix E. All workshops began with the regis- tration and signing in of the participants. After signing in, the participants were given name tags and their individual folders containing material related to the workshop. The duration of each workshop was divided in three large blocks: introduction, interactive workshop, and debrief and discussion. The activities performed in each are explained in Sections 4.2.3.1 to 4.2.3.4 below. 4.2.3.1 Introduction The introduction consisted of three segments: a welcome to the workshop participants, an introduction to scenario planning method, and an introduction to the infrastructure segments chosen for the workshop. • Welcome: A high-level executive at the host organization (such as Ms. Paula Hammond at WSDOT or Mr. Bill Gardner at MNDOT) kicked off the workshop with a welcome. Besides formally welcoming the participants to the workshop, the hosts also noted that their organi- zations were planning to bring the results of the workshop into their planning process. This speech was an endorsement to the gravity of the workshop. While most workshops had a high-profile official from the planning organization give this speech, not all did due to different levels of involvement of the agencies. The lead facilitator gave the welcome address at POLB (Long Beach) and GDOT (Atlanta) workshops.

54 Scenario Planning for Freight transportation infrastructure investment • Introduction to Scenario Planning: In all six workshops, the lead facilitator (Dr. Chris Caplice of MIT) introduced the workshop participants to scenario planning. Dr. Caplice used several examples of societal, technological, and political changes that have happened over 20 to 30 years to show that the world we live in today was far different from the 1980s, and could not have been predicted 30 years ago. He also used examples of forecasts being egregiously wrong, to make a case for using scenario planning. This presentation evolved over the series of workshops, and was generally the highest rated part of the exercise. This presentation lasted between 30 and 45 minutes. The PowerPoint slides are part of the Scenario Planning Toolkit and are available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). • Introduction to Freight Infrastructure Segments: In this part of the introduction, the infra- structure segments to be evaluated in the workshop were illustrated to the participants. The goal of doing this was to ensure that all participants had a uniform understanding of what the segments meant. Where applicable, the segments were shown on maps of the region. In the MNDOT workshop, where FABs were used instead of infrastructure segments, this section described the FABs. The introduction to segments or FABs generally took about 15 minutes, and was done by either the planning manager (such as Ms. Barbara Ivanov at WSDOT) or the lead facilitator. At the conclusion of this introductory section, the group took a 15-minute break before partici- pating in the interactive workshop. Before breaking, the lead facilitator informed the participants of the “rules of engagement” in the interactive workshop: no questioning of the scenario, no criticism of ideas, and free sharing of insights. 4.2.3.2 Breakout Sessions For the individual, small group breakout sessions, the participants broke into their pre-assigned groups—one per scenario. The purpose of this exercise was to identify how the planning organization should prioritize its investments in the region’s freight infrastructure in order to be prepared for each scenario. The interactive workshops consisted of several segments lasting from 15 to 30 minutes each. The segments of this session were always performed in the same order, as mentioned below. • Scenario Immersion: At the beginning of the interactive workshop, the scenario facilitators asked their groups to “live in the year 2037” in their scenarios. The participants were then asked to describe the scenario. (All participants were sent the scenario and asked to read it before the workshop.) All scenario facilitators had lists of important facets of their respective scenarios, and facilitated the discussion so that the group would identify most, if not all, of those features. Following this, the scenario facilitators showed the scenario videos (Section 4.2.1.4). The four “Future Newscast” videos are part of the Scenario Planning Toolkit and are available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). The goal of the immersion was to ensure that all participants had understood their scenario well, so they could judge the usefulness of investing in various segments or identify the initiatives for the scenario. Scenario immersion took approximately 30 minutes. • Scenario Implications: After the immersion, the participants were asked to identify the implications of the scenario for the region’s freight infrastructure. The implications were of five types: origin, destination, volume, value density, and the transportation mode of the freight originating, coming into, and passing through the region. Identification of implica- tions took approximately 15 minutes. • Individual Voting: After immersion and identifying implications, the participants were asked to work individually and answer how freight infrastructure funds should be invested today to prepare for the scenario. The participants first wrote their answers individually on the “Individual invest decisions form” (see Section 4.2.1.5) and then placed voting chips on the

Scenario Planning Workshop Design 55 “group voting sheets.” The voting exercise took about 20 minutes. The group took a brief break following the voting, during which the scenario facilitator tallied the group’s vote. Tem- plates for collecting these votes are part of the Scenario Planning Toolkit and are available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). In the “visioning” exercise conducted at MNDOT, the participants wrote their ideas for initiatives in different FABs on sticky notes and posted them on a flip chart. The group and the scenario facilitator summarized the ideas to reveal common themes. • Group Discussion and Consensus: The scenario facilitator then discussed the group’s vote, to identify the reasons why the group had voted as it did. In the workshops where used, the note takers (Section 4.2.1.1) captured the insights shared by the workshop participants. After the discussion, the participants were allowed to change their votes. The group discussions typically lasted for about 20 minutes. • Identification of Initiatives: If a workshop involved a “visioning” exercise after “evaluation,” the participants were asked to identify specific initiatives the planning agency should take today to prepare for the scenario. For this qualitative exercise, the participants wrote their ideas on sticky notes and placed them on a large pad. The group and the scenario facilitator read the notes to identify the common themes for initiatives suggested by the group. The initiative identification exercise lasted for about 30 minutes. The interactive workshops lasted from 1 hour and 45 minutes to 2 hours and 30 minutes. At the end of the workshop, the scenario group was asked to choose two representatives to share the results with the entire group of workshop participants in the “debrief and discussion session” (Section 4.2.3.4). The scenario representatives captured the insights from the group discussion, and the groups broke for lunch. A Facilitators Guide is part of the Scenario Planning Toolkit and is available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). 4.2.3.3 Summary of the Breakout Sessions During lunch, the facilitators tabulated the output of each scenario group (votes, initiatives, insights, etc.) in spreadsheets prepared for the analysis. The charts generated by these spread- sheets were linked into the presentation used for sharing the results of individual scenarios as well the cross-scenario analysis. MIT facilitators prepared the presentations to be shared with the group in the “debrief and discussion session” after lunch. The Excel file used to tabulate and generate results is part of the Scenario Planning Toolkit and is available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). 4.2.3.4 Debrief and Discussion After lunch, all scenario groups assembled together as one group, as they did for the intro- duction session (Section 4.2.3.1). In the first workshop (DVRPC), the debrief session was held over lunch and lasted for only one hour. It was realized that the debrief session needed much more time and focus. Therefore, the debrief and discussion sessions in all subsequent workshops were held after lunch and lasted from 1 hour and 30 minutes to 2 hours. This session typically had the following four components: • Scenario Reveal: The debrief sessions began with the revealing of all the scenarios used in the workshop to the participants. Until this time, each participant had known one and only one scenario—the one he or she participated in. The reason for revealing all the scenarios for the debrief session was to ensure that all participants got to know the different scenarios so as to form a basis for discussion within the group. The scenario videos (Section 4.2.1.4) were used for this purpose. The scenarios summary video containing the overviews of each of the

56 Scenario Planning for Freight transportation infrastructure investment four future scenarios is part of the Scenario Planning Toolkit and is available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transporta- tion Infrastructure Investment”). • Presentation of Individual Scenario Results: After revealing all scenarios, the cross-scenario facilitator (Section 4.2.1.1) invited the representatives from each scenario to present their group’s results. The scenario representative used the presentation MIT facilitators had prepared during the lunch hour (Section 4.2.3.3). The cross-scenario facilitator encouraged the participants in other scenarios to compare their findings to the ones being presented. These often resulted into a lively exchange of ideas among the group. • Presentation of Cross-Scenario Analysis: Following the presentation of results from indi- vidual scenarios, the cross-scenario facilitator presented the charts comparing the results from all scenarios. These charts were used to show the participants how different infrastructure segments can be classified into robust and contingent segments. The Excel and PowerPoint files for this debrief session are part of the Scenario Planning Toolkit and are available on the companion DVD package or online at www.trb.org (search for “Scenario Planning for Freight Transportation Infrastructure Investment”). The classification developed by MIT’s Supply Chain 2020 research group was used for this analysis, as follows: – No-brainer segments are the ones found to be favorable for investment in all scenarios. – No-gainer segments are those that are unfavorable for investment in more than one scenario and not found to be favorable in any. – No-regret segments are those that are favorable in some, but not all, scenarios and are not unfavorable in any scenario. The above three types of segments are robust investments, that is, the decision to invest in them remains the same regardless of the scenario. – Contingent segments are those that are favorable in some scenario(s) and unfavorable in some others. The decisions to invest in these segments are contingent upon which scenario the world comes to be like. The strategy to invest in these segments involves making some “flexible” investments that can be adapted once the planning organization gets a better sense of which way the future may evolve. • Sensors in the Ground: The cross-scenario analysis was followed by a presentation of “sensors in the ground.” Sensors are those events in the business environment that change the assessment of the subjective likelihood of the future evolving in the direction of a particular sce- nario. Sensors provide indications for if and when flexible options in the contingent segment should be exercised. The lead facilitator always gave this presentation. After the discussion of sensors in the ground, the lead facilitator wrapped up the workshop. 4.2.4 Post-Workshop Activities One day after the workshop, the workshop participants were sent a post-workshop survey via email. This survey had two parts. In the first part, the survey asked the workshop participants to assess the usefulness of investing in the freight infrastructure segments used in the workshop. These questions were identical to those in the pre-workshop survey. The reason for asking these questions again was to see if and how the participants’ evaluations had changed after participating in the scenario planning workshop. In the second part, the participants were asked to rate the effectiveness of various parts of the workshop (described in Section 4.2.3) and the material used to describe the scenarios. 4.2.5 Summary This section described the scenario planning process used at the six workshops. The section first presented the resources used in the workshop (Section 4.2.1). This was followed by a detailed account of the activities performed before (Section 4.2.2), during (Section 4.2.3), and after (Section 4.2.4) the workshop.