Development and Validation of a Seven-County Regional Pedestrian Volume Model

Abstract

This study describes the development and validation of pedestrian intersection crossing volume models for the seven-county Milwaukee metropolitan region. The set of three models, among the first developed at a multi-county scale, can be used to estimate the total number of pedestrian crossings per year at four-leg intersections along state highways and other major thoroughfares. Outputs are appropriate for annual volumes ranging from 1,000 to 650,000. We used negative binomial regression to relate annual pedestrian volumes at 260 intersections to roadway and surrounding neighborhood socioeconomic and land-use variables. The three models include seven variables that have significant positive associations with annual pedestrian volume: population density within 400 m of the intersection; employment density within 400 m; number of bus stops within 100 m; number of retail businesses within 100 m; number of restaurant and bar businesses within 100 m; presence of a school within 400 m; and proportion of households without a motor vehicle within 400 m. Results suggest that square root or cube root transformations of continuous explanatory variables could potentially improve model fit. The models have fair accuracy, with each of the three model formulations predicting 60% or more of validation intersection counts to within half or double the observed value. Future research could address overprediction by creating new variables to better represent the number of lanes on each intersection leg and low socioeconomic status of adjacent neighborhoods.

Comparing Direct Demand Models for Estimating Pedestrian Volumes at Intersections and Their Spatial Transferability to Other Jurisdictions

“…The applications for Milton are the ones that provided the best performance. The ratios of MAE and average true AADPT ranged from 0.52 to 0.78 and are similar in magnitude to those reported for the original calibration/validation data sets of the DD models (8,17). We believe that there are two reasons for this good level of spatial transferability.…”

“…Furthermore, four-way intersections and those located on major or minor arterials are associated with more pedestrian trips than three-way intersections and those located on collector streets, respectively. Similar tables to Table 2 can be found in Singleton et al ( 15 ), Schneider et al ( 17 ), and Munira and Sener ( 18 ).…”

“…For example, the presence of a commercial establishment or a school might produce fewer pedestrian trips in Pima County than in the sites from the original calibration. The Schneider et al ( 17 ) model performed the best in Pima County and is also the one with the lowest average AADPT. Different relationships between explanatory variables and the number of produced trips (i.e., model coefficients) across jurisdictions may be related to jurisdictional specific characteristics such as level of motorization, incentives to use active transportation, land-use mix, weather, and others.…”

“…Overall, the models developed by Singleton et al (15) and Sanders et al (7) presented the best and worst transferability performance, respectively. The model proposed by Schneider et al (17) provided reasonable accuracy for Milton, the best performance for Pima County, but showed unrealistic estimates for Toronto. The model calibrated by Hankey et al (6) consistently overestimated for Pima County and The key takeaway from this paper is that the DD models evaluated in this research performed considerably differently across jurisdictions with distinct land use and socioeconomic features and levels of pedestrian activity.…”

“…Two strategies have been considered to deal with this kind of data: a) taking the arithmetic average of the variable for every zone that is intersected by the buffer; and b) taking the weighted average of the variable based on the buffer intersection area. Some authors explicitly identified which strategy they followed: Schneider et al ( 17 ) used an arithmetic average, and Hankey et al ( 6 ) applied a weighted average. When no information is provided, the weighted average is assumed.…”