HERS-ST Discussion
[ Return to Subject List ]
Messages posted for Performance Measures
Subject: Performance Measures
From:
Rich Arnold
Date: 10/24/2003 |
| In evaluating a process for developing Sensitivity Testing for HERS-ST, it is clear that I need to define and develop some operating characteristics or performance measures that will be useful in evaluating levels of credibility for various data elements. It is probably a no brainier that the lane and shoulder width data are not as important in the overall analysis as the AADT and percent truck. However, how close do I need to be with the AADT? Within 5%? Within 10%? How close do I need to be with the percent truck in order to be comfortable with the analysis? How will a 2% change in trucks effect the HERS-ST analysis? The numbers or costs of improvements identified by the model are not enough to determine a level of confidence in the data quality issue. In the past, I have used both PSR and AES values to track performance over time (across the 20- year analysis period), but am wondering what additional types of measurements might be useful and/or available for tracking the highway performance. I expect that we will need to decide what is being measured before we can know whether we can get close enough with the data that is available. Any thoughts on the subject???? |
Subject: RE: Performance Measures
From:
Susan Gorski
Date: 10/27/2003 |
| Rich, The question I think is more in indepth than HERS-ST can do. You are now treding into life cyle cost analysis. If you are looking at congestion, you need to look beyond the 30th high hour (DHV) and asses duration. Also for congestion on non freeways, lane, shoulder width percent sight restriction play do affect the LOS. Pavement condition requires not only the number or percent of trucks but the KIPS. |
Subject: RE: Performance Measures
From:
Herb Weinblatt
Date: 10/28/2003 |
| Rich, Some performance measures that can be used to supplement or replace average speed are Incident Delay, Other Congestion Delay, Travel-Time Costs (per 1000 VMT), and Pct. VMT on roads with peak- hour V/C above some user-specified threshold (but note that this represents all VMT on these roads, not just VMT occurring when the road is congested). There is no single answer as to how close you need to be for any given variable, but, for any variable, you can get some idea by performing a sensitivity analysis; e.g., by increasing AADT by 10% and seeing what kind of effects you get. For pavement analysis, HERS incorporates FHWA- developed default values of average ESALs per vehicle by functional system, but these can be replaced (in advanced mode) with your own values. However, HERS has no capability for using site-specific values. Also, Pavement Mgmt. Systems (which are specifically designed for this purpose) can be expected to provide substantially better pavement analyses than HERS. |
|
