jrawbits
commented
2 years ago On issue 1, John is right. The PCE for capacity is a linear function of vehicle size and required following distance, whereas the "damage equivalent" is an exponential function of vehicle weight. We need to be clear that inputs are being appropriate described and used relative to roadway damage and congestion (both are considered).
On issue 2, we need to do a deep dive into what BalanceRoadCostsAndRevenue is actually doing, relative to what the documentation says it should do.
Issue 1. Within the file region_road_cost.csv, the sixth attribute—passenger car equivalents per heavy truck or shown as HvyTrkPCE—may not be serving its intended purpose. The VisionEval documentation notes that PCE is the “number of light-duty vehicles a heavy truck is equivalent to in calculating road capacity.” In terms of congestion, this factor is correct—the Highway Capacity Manual uses a similar figure when converting trucks to passenger cars for the purposes of determining lane capacity. However, the R script seems to suggest that this factor is used to convert truck VMT to passenger VMT in terms of roadway damage. If that is the intention, a much larger factor than 3 is probably appropriate; for instance, older research suggests a factor of about 93.25 where, in 1990, over a one year period, an increase of one vehicle for a one mile roadway section increased maintenance costs by 4 cents compared to an increase of one truck that increased costs by $3.73. (That older research is here: http://onlinepubs.trb.org/Onlinepubs/trr/1990/1262/1262-006.pdf)
Issue 2. The BalanceRoadCostsAndRevenues script explains that if user fees are not large enough to cover the cost of roadway maintenance, then an extra tax will be levied to make up the shortfall. That extra tax is called “ExtraVMTTAx” and is an output field attribute in the Region_2045.csv file. However, the value of this attribute is always zero--even when all values in region_road_cost.csv are multiplied by a factor of 10! Thus it may be the case that this file is used but perhaps not in a different way than is presently understood. (To be clear, the file must be included or else VisionEval will not run.)