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.
trips from within the North Corridor to the CBD are underrepresented by 5%. Regionally, the model is over- representing trips to Ohio State University (OSU) by just 3%. There are specific travel markets that are weak, including a 27% underestimation of tours from the North Corridor to OSU. Work-tour productions and attractions are well estimated by the model. Almost all markets are represented within 10% of the CTPP totals. Figure 2 shows the modeled travel analysis zones with work-tour destinations in the CBD. Figure 3 shows the same for census blocks from the CTPP. While the zones donât match one- to- one, patterns are reflected fairly accurately. The near northwest and the second ring east are two high- income suburbs that are reflected in both maps. The area around the southern circle is OSU, and the small size of the zones in that area potentially obscures the correlation there (also note the above dis- cussion). The model smoothes the employment in the near northeast more than is shown from the CTPP. The far southwest shows a high amount of CBD- oriented workers, while eastern Delaware County shows a fair number of CBD- oriented workers. HIGHWAY ASSIGNMENT VALIDATION Model validation refers to the comparison of estimated and observed individual highway link loadings and tran- sit route boardings. The purpose of model validation is to gauge how accurately the model predicts observed base- year travel patterns and to identify potential model shortcomings. The MORPC model was validated against traffic counts that have been processed to represent directional average annual daily traffic for the year 2000. The criteria used to assess the adequacy of the model val- idation were: percent vehicle miles traveled (VMT) error, percent VMT root- mean- square error (RMSE), and per- cent volume RMSE, by facility type and volume group. Highway assignment validation was geographically structured by districting schemesâ rings, sectors, and super districts. The validation by volume group is shown in Figure 4 and Table 4. All volume groups, except 0â500, fall below the maximum allowable percent RMSE. (Maximum Allowable %RMSE per ODOT Traffic Assignment Pro- cedures, page 30.) Table 5 shows validation statistics by facility type. Total VMT is within 1% of the observed data, and total volume is within 2% of the observed volumes. With respect to the geographic districts as shown in Figure 5, the results demonstrate a validated model with respect to observed counts by each of the three district- ing schemesâ concentric rings, radial sectors, and super districts. 168 INNOVATIONS IN TRAVEL DEMAND MODELING, VOLUME 2 FIGURE 2 2000 model work trips to CBD. FIGURE 3 2000 CTPP work trips to CBD. 250 Pe rc e n t 200 150 100 50 0 1 0â 49 9 2 50 0â 1,4 99 3 1,5 00 â2 ,49 9 4 2,5 00 â3 ,49 9 5 3,5 00 â4 ,49 9 6 4,5 00 â5 ,49 9 7 5,5 00 â6 ,99 9 8 7,0 00 â8 ,49 9 9 8,5 00 â9 ,99 9 10 1 0,0 00 â1 2,4 99 11 1 2,5 00 â1 4,9 99 12 1 5,0 00 â1 7,4 99 13 1 7,5 00 â1 9,9 99 14 2 0,0 00 â2 4,9 99 15 2 5,0 00 â3 4,9 99 16 3 5,0 00 â5 4,9 99 17 5 5,0 00 â1 20 ,00 0 Model % RMSE Max % RMSE FIGURE 4 % RMSE by volume group.