Spatial Analysis Of Large Scale Freight Commodity Survey Data For Systems Planning

The Ministry of Transportation Ontario (MTO) Systems Analysis and Forecasting Office (SAFO) presented a case to determine if there is a relationship between the nationally collected Trucking Commodity Origin and Destination (TCOD) survey data and the provincially collected Commercial Vehicle Survey (CVS) data. The MTO performs the CVS every five to six years across the province of Ontario. It is conducted by roadside surveyors at 150 locations. The survey is very costly and requires a substantial amount of time and resources to complete. Though the CVS collects a large amount of trucking data, more information is required to gain a better understanding of freight movements within the province. The TCOD survey is a more comprehensive survey with more data points. The survey data is collected via phone interviews, electronic data reporting and on-site visits to shipping companies. A relationship between the two databases could allow for TCOD survey data to be used to populate the CVS database with additional information, without the costs associated with performing a CVS. In this Master of Engineering project, raw data collected by both CVS and TCOD surveys has been aggregated on municipal and zonal levels with the purpose reducing the size of the databases to include only the Greater Golden Horseshoe (GGH) area and to compare the characteristics of the two databases. The TCOD database contained information for all data collected in Canada, with 215,001 data records. The CVS database contained all freight information for Ontario, with 10,758 data records. To reduce the database sizes, ArcGIS was used to link the locations of data points to the municipalities and transportation assignment zones in the GGH. The output from ArcGIS listed all locations with associated municipal and zonal identification numbers. This information was linked to the TCOD and CVS databases using Microsoft Access, resulting in a complete table of locations, identification numbers, municipality names, trucking company type, truck weights, and commodity type within the GGH. Density maps were created to provide a qualitative assessment of the two surveys. This demonstrated that most of the trucks that were surveyed were either originating or arriving in the Greater Toronto Area (GTA). The CVS highest daily weights were located in Toronto, Mississauga and Hamilton. This is expected as these municipalities are three of the largest economic centers in Ontario. The TCOD data follows the same trend where Toronto, Mississauga and Hamilton are at the top of both origin and destination highest density daily weights. Though the density trends are similar, the TCOD survey differs from the CVS because it has more data for the outer regions of the GGH. This is expected because the CVS is only performed at a limited number of roadside locations while TCOD uses phone, mail and visits to shipping companies to provide extensive coverage of the GGH. The effectiveness of the CVS site locations was evaluated with a point density spatial analysis. All CVS origin and destination weight values were plotted on the GGH map and centres of high densities were identified with dark circles. These locations were not restricted by an assigned municipal zone, allowing the natural centres of high densities to be identified. The centres do not always falls within municipal boundaries, which indicates that evaluating truck activity centres from a municipal perspective may not provide a true representation of where the high freight activities are located. Based on the high density points found in this analysis, the existing CVS sites appear to be positioned in ideal locations which may provide good coverage of these freight activity centres. A commodity distribution comparison was performed for the overall weight values for CVS and TCOD surveys in the GGH. The distributions had some similarities, but also varied in some areas. Additional commodity distribution comparisons were performed for three cities; Toronto, Mississauga and Hamilton. Overall, the commodity distributions for the CVS appeared to be more evenly distributed and consistent than the TCOD survey. This may indicate that the TCOD should improve its effort toward the less represented commodity types. Although this study is thorough for areas within the GGH, the results could improve if both data sets were collected in the same year; in this case CVS data was collected in 2006 and TCOD data in 2010. During this time, the freight movement trends may have changed, particularly because of the 2008 economic recession. All aspects of the economy were impacted during this time, and the effect on the shipment of goods must be kept in mind when conducting comparative studies. This analysis only considered data points within the GGH. This may have affected the results, as it is possible that a stronger relationship between the databases may exist outside the GGH. An evaluation of freight movements throughout the Province of Ontario is recommended. The municipality density maps, point density maps and commodity distribution analysis indicate that the coverage between the two surveys show some similarities, but overall are inconsistent. In some cases, the CVS had more data points for a particular municipality or commodity; while other times the TCOD survey had more data. This may indicate that finding a relationship which produces a conversion factor that can be applied to the TCOD database to populate the CVS database may be challenging. However, this analysis confirms that the CVS has much fewer data points than the TCOD survey, for the majority of the municipalities in the GGH. Finding a potential relationship between the two surveys would be very beneficial to the MTO as it could provide a more complete database of freight movements in Ontario. A regression analysis between the two databases is recommended as it may be able to identify a potential conversion factor between the CVS and TCOD databases.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.