Guide to Effective Methods for Setting Transportation Performance Targets (2022) / Chapter Skim
Currently Skimming:
Pages 21-38
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 21... ...
21 Safety performance reporting to FHWA occurs annually, rather than on the biennial reporting cycle of other national measures. The safety performance measures cover fatal and serious crashes that occur on all public roads.
|
From page 22... ...
22 Guide to Effective Methods for Setting Transportation Performance Targets Worsening Performance and Targets A disinclination to set worsening targets (i.e., targets reflecting increases in fatalities or serious injuries) is present across all performance areas to some degree, but this sensitivity is particularly acute for safety performance measures, given that people's lives are literally at stake.
|
From page 23... ...
Target-Setting Methods for Safety 23 Method Strengths Limitations Other considerations Targeted reduction: A defined decrease from the baseline, often based on policy or long-term vision goals Simple, easy to communicate; in line with agency's aspirations No insights into causes of outcomes -- Time-series trend: A simple, univariable forecast based on historical trend data Simple while still being data driven No insights into causes of outcomes. May result in a worsening target Trend plus other factors: Manual adjustment made to forecast results to account for other considerations Begins to bring in prominent influences on outcomes Adjustments might be data informed but may stem from other motivations May still result in a worsening target, though the agency has more ability to limit this Multivariable model: Regression or time-series model that incorporates explanatory variables to predict performance Fuller understanding of factors associated with outcomes, which thereby informs decisions Complex, time needed for data gathering, requires analytical and data skills, harder to communicate May result in a worsening target Table 2.
|
From page 24... ...
24 What It Is A targeted reduction refers to a defined decrease from baseline conditions without specific regard to past trends. This is often in the form of a set annual decrease (e.g., 2% each year)
|
From page 25... ...
Target-Setting Methods for Safety 25 Step 2: Identify Goal Following the conversations with stakeholders, identify the goal level around which to coalesce target setting. Agencies may have multiple long-term goals that could form the basis for targets, such as a Vision Zero goal or a long-term goal in the SHSP, and discussions might center on which of these to use as the basis for targets.
|
From page 26... ...
26 Guide to Effective Methods for Setting Transportation Performance Targets The Louisiana HSP has a chart, show in Figure 1, that depicts the state's straight-line targeted reduction of 1% per year. The Florida DOT has a firm target of zero for all federal safety measures.
|
From page 27... ...
27 What It Is A time-series trend forecast relies solely on historical performance data as the basis for the projection and eventual target. A trend is established for the most recent period of performance by using either a straight-line or other built-in function (e.g., logarithmic, exponential)
|
From page 28... ...
28 Guide to Effective Methods for Setting Transportation Performance Targets serious injuries. Nonmotorized crash data are the least established, and this measure often relies on fewer than 10 years of data.
|
From page 29... ...
Target-Setting Methods for Safety 29 Examples The Mississippi DOT used data from the period 2009–2018 for fatalities and nonmotorized fatalities and serious injuries and from the period 2009–2017 for serious injuries to establish a trend of 5-year averages. The agency applied Excel's FORECAST.ETS function to these data (FHWA 2020)
|
From page 30... ...
30 What It Is In many cases, agencies decide to adjust the results of their quantitative projection to establish the final target. This can be done on top of any of the methods discussed above, though it is most commonly applied on top of a time-series trend.
|
From page 31... ...
Target-Setting Methods for Safety 31 How to Do It Step 1: Gather Stakeholders and Initial Target Results This method of target setting begins with the results of any of the previous methods, and in fact can include the results of multiple methods to be discussed and considered together. Identify the stakeholders whose input is sought on the final target level.
|
From page 32... ...
32 Guide to Effective Methods for Setting Transportation Performance Targets Limitations This method may move targets away from being truly data-driven by incorporating influencing factors in a nonquantitative way that could include human error or bias. Examples The New York State DOT employed a linear trend line approach that used the 5-year average as the data point for each year to develop its 2021 targets for the number of fatalities, the number of serious injuries, and the total number of nonmotorized fatalities and serious injuries.
|
From page 33... ...
Target-Setting Methods for Safety 33 rates of serious crashes. While it is more difficult to assess crash reductions from education and enforcement programs, a percentage annual reduction in fatal and serious injury crashes was also determined for these programs.
|
From page 34... ...
34 What It Is A multivariable statistical model is a method that attempts to quantify forecasts in a mathematical framework while incorporating exogenous explanatory variables to improve accuracy and connect results to other key factors. There is a wide array of forms and options for multi variable statistical models, including multiple regression analysis or multivariable time series.
|
From page 35... ...
Target-Setting Methods for Safety 35 use was the time series model ARIMA with unemployment as a covariate. ARIMA can be run as a simple univariable model or include covariates as explanatory variables, in which case, it may be referred to as an "ARIMAX" model.
|
From page 36... ...
36 Guide to Effective Methods for Setting Transportation Performance Targets The selection of model order will depend on the user's data and variable count. Selecting a model order can be done by using domain knowledge and reasoning as well as by using automated tools available in your statistical package.
|
From page 37... ...
Target-Setting Methods for Safety 37 Step 4: Compare and Confirm Model Fit To determine the optimal model form for a given situation, practitioners should review each prospective model's fit and error metrics, where lower error metrics indicate a better model fit. Both the Akaike information criterion (AIC)
|
From page 38... ...
38 Guide to Effective Methods for Setting Transportation Performance Targets The Virginia DOT changed its target-setting philosophy in 2017 when the agency was seeing increases in fatalities and serious injuries and there was no way to hit its ambitious 5-year projections. The agency's assessment was that the targets were too optimistic to be useful, and it needed to have more detailed predictions from which to make decisions.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.