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.
Final Report Project No. 08-94 60 CHAPTER 4. SUMMARY AND RECOMMENDATIONS 4(A) SUMMARY The NCHRP 08-94 research study produced three products: ï· An online software product called TFGuide to assist travel forecasting practitioners in selecting travel forecasting methods to support planning programs. This product is a stand-alone product that includes an online reference guide and a full suite of help features to assist the user. ï· A travel forecasting User Guide for TFGuide that includes reference guides for all information contained in the software tool. ï· A Final Report that documents the research approach to develop TFGuide, the administrative features, and the pilot testing of the software. This Final Report was developed separately from the User Guide so that the Final Report could include the background research, software design, and administrative functionality, whereas the User Guide more directly supports the use of TFGuide. The state-of-the-practice review was conducted to identify the most commonly used methods for travel forecasting and to identify references that are useful for each of these methods. This review focused on travel demand models (traditional passenger, advanced passenger, transit passenger, and freight and goods movement models) and traffic assignment and simulation models (integrated multiresolution, static UE assignment, DTA, and traffic microsimulation models). The TMIP agency needs assessment was a source of useful information about practitionersâ planning needs and methods. The software design prioritized flexibility, providing users with ways to prioritize requirements and metrics and constraints in selecting a method and in building the technical content and relationships in the system. This flexibility is provided via the administrative functionality so that the system can be updated and enhanced. This structure also easily accommodates additional methods (and programs and requirements and metrics) so that future versions of the software could include new technical content without any major structural changes to the software. This administrative function allows technical experts to develop and review new content before updating the software. The administrative functionality also provides flexibility in establishing relationships for each method, program, requirement, and metric in the system. 4(B) RECOMMENDATIONS FOR FUTURE ENHANCEMENTS The scope for the NCHRP 08-94 project focused on travel forecasting methods directly, but travel forecasting practitioners are faced with challenges that require additional methods to support travel forecasting. These additional challenges called upon methods (Figure 32) for economic, land-use and demographics, person and household characteristics, impacts and performance metrics, project prioritization, and visualization. The research team worked with the
Final Report Project No. 08-94 61 expert panel to suggest these additional methods for future versions of the software to focus the development on the primary travel forecasting methods first. TFGuide was designed for an administrator to maintain and update the software. Given the dynamic nature of the travel forecasting industry, maintenance and updates will allow TFGuide to be refined with user input, expanded to include newer methods, and updated so the technical content can remain current. The administrator will manage user accounts, provide technical support, and receive user feedback on the recommendations. The administrator could also review user scenarios for credibility and reliability of the recommendations. The state-of-the-practice review, pilot tests, and feedback from the expert panel also identified new features that can be considered for future versions. The TMIP needs assessment survey identified the three most important challenges facing travel forecasting practitioners: ï· Limited staff time for planning analysis. ï· Limited budgets for planning analysis and the high costs of planning analysis. ï· Lack of data or poor data quality. Staff time could be added as a current resource under Expertise, but the staff time required for these planning analyses could vary significantly depending on whether the work is completed with agency staff or consulting staff. If this feature is desired, then an additional user input could be whether agency staff are managing or completing the work in-house. TFGuide can provide information on budgets needed for planning analysis and suggest more cost-effective methods. TFGuide currently identifiesâin broad termsâthe types of data (and capital) required to support recommended methods, but does not provide any details on these data sources or an assessment of data quality. If this were desired for a future version, then it would be a significant enhancement to include necessary technical details and data quality assessments for each data source.
F Final Repor IGURE 32: FUTU Economic, Use an Demograp ⢠Input-Output Mod ⢠Structural Equilib Models ⢠Evolutionary Part Equilibrium Mode ⢠Evolution Models firms) ⢠Land Use Visionin ⢠Land Use Foreca ⢠Population Synthe ⢠Firm Synthesizer ⢠Household Class Models ⢠Integrated Land U Travel ⢠Delphi Process (n model) t RE METHODS FO Land d hic els rium ial- ls (pop and g sting sizer ification se and on- ⢠W ⢠To ⢠Dr ⢠Ph ⢠Ve ⢠Ve ⢠Bi R CONSIDERATIO Person and Household Characteristic orkplace/School Loc ll Transponder iver's License ysical Mobility hicle Availability hicle Type Choice ke Ownership N IN THE METHO s ation Im Pe ⢠Capaci ⢠Safety ⢠Benefit ⢠Air Qua ⢠Strateg ⢠Econom ⢠Health ⢠Equity/ Justice ⢠Sketch ⢠Growth Analysi ⢠Perform 62 D SELECTION TO pacts and rformance Metrics ty and Intersection (Reliability) -Cost lity ic Planning Models ic Impact Impact Environmental Analysis Planning Models Factoring/Trend s ance Monitoring OL Pro Priorit ⢠Scorecards ⢠Benefit-Cost ject ization (non-model) Methods Project Visualiza ⢠Spatial Data Map ⢠Statistical Chartin ⢠Tabular Summar ⢠Inforgraphics, Min And Data Schem ⢠Multimedia & Inte (which technically with each of the o ⢠Illustrations & Re No. 08-94 tion ping g ies dmaps, atics raction overlaps thers) nderings
Final Report Project No. 08-94 63 The expert panel and pilot test agencies also identified several features that would be useful to consider for future versions of the software: ï· TFGuide is limited to considering one planning program at a time. Planning agencies would like to consider methods that support multiple planning programs; expanding TFGuide to consider multiple planning programs at once would be helpful. ï· TFGuide allows weights on a performance measure category, rather than on individual performance metrics. Expanding the weights to include individual performance metrics would be a useful enhancement. ï· TFGuide does not filter responses for performance metrics based on earlier choices for planning programs. This feature would make selections for performance metrics easier, but may limit users who are trying to include additional metrics for an individual program. At this point, the software is designed with this flexibility in mind, but could be modified to constrain some choices for performance metrics. The research team has provided a user account for general purposes, but user accounts will need to be created so that each user can develop and maintain her or his own scenarios, along with current methods and resources. These user accounts will be managed by the administrator of the software. The software can be distributed on TFResource or some other suitable website.
