Knowledge CentreTechnical Resources SearchConference PapersBig Data Analysis to Measure Delays of Long-haul Truck Trips

Big Data Analysis to Measure Delays of Long-haul Truck Trips


Transportation delays are a commonplace occurrence for road users where congestion can cause substantial stress and loss in time for passenger car users. On the other hand, delays have a direct impact on a region’s economic performance due to the added costs incurred by firms. This burden is amplified when delays are unpredictable, resulting in an uncertain arrival time for truck deliveries. The delays associated with truck deliveries can be attributed to a number of factors including congestion on major highways, custom clearance at border facilities, or unexpected events such as road accidents or inclement weather. To understand the nature of truck delays in North America, it is important to determine the time spent delayed in congested highway traffic and at the border if a truck is exporting goods between Canada and the US. Analysis of delays could help firms employ certain strategies (e.g. time-shifting) to optimize their trip duration given the time sensitive nature of goods movement.
This paper seeks to address the prevalence of delays with an emphasis on long-haul truck trips in North America. The analysis makes use of big data that represent the movements of over 56,000 Canadian owned trucks in the year 2013. These movements are captured with approximately 1.1 billion GPS pings. These pings track the geographic position of trucks at different points in time. Numerical algorithms are devised to identify the trips conducted by the trucks and calculate their travel time from the origin to destination. International trips crossing the Canada-US border at several major crossing locations in Ontario are examined first to determine the delays occurring at the border and the relative impact of this delay on the total trip. Next, the total delays observed for truck trips in the dataset are then calculated and compared against the optimal travel time in the absence of delays. Expected delays will be analyzed for different periods of the day based on the average travel time for trips between a given origin and destination to capture the impact of peak and off peak travel on truck movements. Finally, unexpected delays make up the remaining portion of delay that exceeds an average travel time. The results from the analysis will contribute to the ongoing efforts to improve freight fluidity in Canada.

Conference Paper Details

Session title:
How Will "Big Data" Help Us Make Transportation More Efficient
Gingerich, K.
Maoh, H.
Anderson, W.