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.
72 In Chapter 2, we posited the question of why transportation agencies would want to go through all the trouble of using our new tool, based on a complicated SD model, to explore emerging trends and create pos- sible futures. The reason to us is simple: to increase the chance of mak- ing better decisions. But our study delivers more than a new tool. Our goal is to encourage a new way of thinking about alternative solutions to long-range transportation plans for an uncertain future worldâa world that is increasingly more complex and dynamic than the world transportation agencies planned for in the past. Chapter 3 presents eight trends that highlight a range of uncertainties transportation agen- cies face in their long-range planning, from a changing workforce to the implications of a hyperlinked society. In developing Impacts 2050, this study team needed to address some of the limitations of the more traditional approaches to anticipating the implications of future demographic changes. The learning that took place has made it clear that the traditional approaches are no longer sufficient to address the accelerating rate of change and increased com- plexities of the 21st century. Taking full advantage of the products of this study will require individuals who are skilled in the old ways of transportation planning to accept additional ways of thinking, which are more aligned with the 21st-century types of problems and environments. In addition, they need to accept that these new ways of thinking about alternative solutions to long-range transportation plans will sometimes be introduced by individuals junior to themselves, who are more likely to be aware of their potential value. 9.1 Addressing the Dynamic Complexity of the 21st Century What is meant by an accelerating rate of change and increased com- plexity, and how real is it? Almost everyone would accept that today and for some time in the future transportation agencies can count on: ⢠Increased customer/citizen and market/community diversity, ⢠A more transparent worldâsignificantly increased information and knowledge available to customers/citizens, C H A P T E R 9 Opportunities for Improved Decision Making Chapter 9 Takeaways ⢠What is important in todayâs environment is asking the right questions. ⢠Impact 2050 is designed to alter how transportation agencies think about the possibilities of addressing uncertainty and not just deriving the answer to a problem that can be precisely defined. ⢠Reinvention was a design principle built into this research. This recognizes that it is possible to adopt some components of an innovation and change or reject others. ⢠Transportation decision makers must shift from planning strategically to thinking strategically. In this view, formulating and implementing policy become iterative processes and decision making is ongoing and interactive. âThe world moves into the future as a result of decisions not as a result of plans.â Kenneth Boulding, systems scientist
Opportunities for Improved Decision Making 73 ⢠Increasingly more rapid technological change, and ⢠A fiscally constrained operating environment. Agencies can also expect that: ⢠They will have less time to respond to market/constituency requests, ⢠They will be less able to forecast market conditions and constituency needs, ⢠The places they normally go to get money and things done will be under pressure to change, and ⢠The ability to find, much less talk to, mass markets through a single communication medium where they control the message content will continue to decrease. Based on these changing conditions, the approach and the need for change are best described by revising a very important point made in 1960 by John Tukey when he said: Far better an approximate answer to the right question, which is often vague; than the exact answer to the wrong question, which can always be made precise. While this quote has remained relevant for 50 years, due to changing 21st-century conditions we felt the need to update Tukeyâs wisdom with the following revisions. Far better a system of inquiry that engages decision makers and information providers to ensure the appropriate data not only are collected but are also organized in manner that the data can be transformed into relevant answers to the right questions in a timely manner at a cost that decision makers will value. Thus we can avoid processes that imply a precise answer, in some cases, to the wrong question. Long-range planning traditionally uses forecasts generated from travel demand forecast mod- els. These models focus on quantitative accuracy in input data and model parameters. Output is in the format of point estimates for specific years. Running the model and analyzing results is time-intensive. Our solution, Impacts 2050, facilitates analytical reasoning, rather than the com- putation of a precise number. The focus is on including a wide range of model relationships and on âqualitative accuracy.â Running the model is relatively quick and easy, therefore allowing a user (from a planner to a forecaster to a decision maker) to iteratively change model parameters and inputs to gauge the impact on travel behavior. What becomes important in this environment is asking the right questions. 9.2 Changing Oneâs Mindset to Handle Complex and Dynamic Problems In Chapter 4, we provided the rationale for our studyâs solution to the problem of long-range planning in an uncertain world. Our solution reflects an emerging trend in long-range plan- ning, where there is an awareness that one cannot actually forecast the future, but that many scenario possibilities need to be studied, so that a policy or investment strategy that minimizes risk or moves toward some desired goal(s) can be followed. Chapter 5 presents four plausible future scenarios, but with our tool, we invite users to develop their own scenarios. Our scenarios were designed to encourage transportation agencies to consider what they would do if the future took a decidedly different direction over the next 50 years compared with its historical trajectory. From a systems thinking perspective, a joint scenario/modeling approach can replace the tra- ditional process of transportation forecasts, feeding long-range plans with an organic decisionâ a support system that can pump a free flow of contextual data and knowledge into a series of dialogues that take place continuously across the planning functions of transportation agencies. A distinction between two metaphors helps illustrate the importance of these differences. The industrial-age mechanistic mindset encouraged us to think about managing enterprises as if they
74 The Effects of Socio-Demographics on Future Travel Demand were made up of replaceable partsâlike pieces in jigsaw puzzles. The metaphor fit reasonably well for that era. When you start a puzzle, you know how many pieces youâre supposed to have, and the chances are they are all there. Each of the parts will interact with only a small portion of the other parts. If you have trouble deciding how to put the pieces together, you have a picture on the box to remind you there is one single solution to the problem. And, though some puzzles are more complex than others, the underlying process of putting them together is always the same. But todayâs business challenges are more complex than that. Transportation agencies operate in a world of complex problems compounded by an accelerating rate of change. It is an environment that consists of constantly changing processes, relationships, and interacting componentsâmore like a DNA molecule than a jigsaw puzzle. Depending on how the pieces come together and what is occurring in the containing environment of the molecule, we can end up with a final result different from what we had any reason to expect. Impacts 2050 was designed to facilitate a change in the way transportation agencies think about the futureâto focus more on the possibilities of addressing the uncertainty of complex and dynamic problems than on deriving the answer to a problem that can be precisely defined. Impacts 2050 allows transportation planners to engage in a decision-making process that allows them to consider alternative strategies based on an assessment of the impact of alterna- tive future scenarios. Chapters 6 and 7 provide building blocks of the tool and the informa- tion that can be derived from it. These products were built not only to provide estimates of future needs but also to enable users to develop a deeper understanding of the dynamics and inter action of socio-demographic and other influencing factors, rather than providing exact forecasts and predictions that are limited by what is known at the time the model is run. This approach allows the planning team to think about and consider alternatives to the projected estimates, rather than being limited to focus on the implications of the outcome of traditional planning models. 9.3 Reinventing the Model to Meet the Requirements of an Uncertain Environment The basic elements of Impacts 2050 have been designed so they can be modified to adjust to the differences that distinguish one transportation problem for another. Data have already been entered into the tool for the Atlanta, Boston, Detroit, Houston, and Seattle regions. The study team used publicly available data. But because a region or state may have more reliable or accu- rate measures than we provided, flexibility was built into the tool so that the underlying data can be improved. The four scenarios are predefined in the tool. But a user can create new scenarios by altering the scenario parameters. Easy instructions direct a region other than our five (or a state) to set up the model for itself. âReinventionâ was a design principle built into our approach. The concept of reinvention recognizes that an innovation is often really a bundle of components; it is possible to adopt some components and change or reject others. This approach is reflective of the work of Everett Rogers in his original work in the diffusion of the Census Bureauâs geographic systems that served as a precursor to the TIGER system. Rogers and his colleagues identified their findings as reinvention. They defined this important finding as follows: Reinvention is the degree to which an innovation is changed by the adopter in the process of adoption and implementation after its original development. Reinvention may involve both the innovation as a tool and in its use. Thus, the same technological innovation may be put to a different use than originally intended; alternatively, a different innovation may be used to solve the same problem. In addition, the intended or potential consequences of an innovation may be changed through reinvention.
Opportunities for Improved Decision Making 75 9.4 Doing Better The current practice of long-term planning relies on point prediction in an uncertain future. The predictions are updated at certain intervals (every 10, 15, 20 years) in order to correct course. As consumers of forecasts, decision makers face the problem of interpreting the findings they receive. Through our study, we offer a tool and suggest an alternative approach to cope with this uncertainty and to make better decisions. Using the tool requires a change in thinking, in which the output of the forecast is a less important ingredient to a long-range plan than the process of interacting with the model to produce many different possible future scenarios. Chapter 8 provides implications of the use of the tool for transportation agencies and key indica- tors that can be monitored as early warning signs of change. The premise is that transportation decision makers shift from planning strategically to thinking strategically. In this view, and with the support of Impacts 2050, formulating policy and implementing policy become iterative processes, and decision making is ongoing and interactive. 9.5 Getting the Research into Practice This project has been unique and ambitious in terms of its scope: to identify an approach for assessing the impacts of socio-demographic changes on future travel demand and to develop an associated tool to simulate regional changes in population demographics over time, with feed- back from employment, land use, and transportation supply. While the breadth of the relation- ships modeled in the tool approaches the most ambitious of integrated regional transportation models, we have captured the detailed spatial aspects of such models in approximate relation- ships. This is to be able to simulate the trends and feedbacks in response to a wide range of dif- ferent user inputs and assumptions, over an extended period of time. Further, we have built the tool into a spreadsheet-based format that is designed for use by both technical and nontechnical users. That it came close to accomplishing our ambitious goals was evidenced by the positive responses in the initial demonstration of the tool for five different U.S. regions, as discussed in this report. But with a project of this scope, it is inevitable that many further improvements will be made. Potential improvements have been identified from our initial pilot users for this project, as well as feedback from other potential users in regional and state agencies other than our five pilot sites. Below are some recommendations for further development of the tool: 1. Improving the toolâs usability. Building a fairly complex model into a relatively simple and familiar user interface is a challenge. While using the tool, we have identified a number of ways that the functionality, stability, and ease of use could be improved. Some examples are: ⢠Improve the code and installation process to ensure that the tool will run correctly on a wide range of computing systems (including Apple computers), under a wide range of user configurations of those systems. The code would also anticipate a wider variety of user errors in installing and using the tool and would prevent or correct those errors. ⢠Make it simpler to create, arrange, and catalog a wide variety of different user scenarios within the tool. One approach for this is to store the inputs and results for each new sce- nario within the same version of the spreadsheet (rather than saving each one as a new version). Each new scenario would then be added to a menu list, so that the user can select it as the starting point, to either rerun it or modify it to create a new scenario. ⢠Include more functionality for creating and saving custom reports in the form of tables and graphs. ⢠Such changes will require a great deal of expertise in programming Visual Basic macros for Excel.
76 The Effects of Socio-Demographics on Future Travel Demand 2. Refining the transition rates within the demographic sector. The socio-demographic tran- sition rates within the model are based on analysis of the most appropriate available data, including the Panel Survey on Income Dynamics (PSID) and various types of Census data. Yet, there were only small samples sizes for estimating some of these rates, particularly those for immigrant and minority households. Very recently, a new wave of PSID data from 2011 has become available that would increase the useful sample size for analysis, so it would be worthwhile to update this analysis at some point. Detailed analysis of Census micro-data from 2000, the Public Use Micro-data Sample, and the subsequent American Community Survey could also be useful, although the sampling error in repeated cross-sections must be considered carefully in such analysis. 3. Conducting sensitivity testing. The use of model sensitivity testing is common with tradi- tional models. In Impacts 2050, we generally can assess model sensitivity by changing param- eters in the model spreadsheet. However, users in specific regions may ultimately find it useful to document a range of sensitivity tests comparing the sensitivity of selected dependent vari- ables on a different independent set of variables for (1) comparison within the same scenario, (2) between different scenarios, and (3) between different metro areas for the same scenario. 4. Convening a TRB Subcommittee to the Travel Model Improvement Program subgroup centered on the model/scenario planning process. The group would monitor the use of Impacts 2050, identify helpful modifications made by users that would benefit other users, and promote the new planning process at conferences and other events. The group could actively support model enhancements, including continued funding for updating data sources, such as the PSID.
