Highways as the primary type of transportation infrastructure, allow for haulage of huge amounts of goods and services in North America. An aging infrastructure combined with several other parameters such as high volume of heavy truck traffic and long harsh winters could lead to faster pavement deterioration. To improve safety, traffic planning, geometry, and structural performance of the highways, transportation agencies should have a detailed program to maintain, preserve, and reconstruct these infrastructures. Workzones interrupt the regular traffic flow on the highways which could potentially lead to an increase in the number of collisions. To prevent this, agencies are encouraged to present a comprehensive strategy to minimize the queuing in the workzones. The Ministry of Transportation Ontario (MTO) have identified that accurate prediction of workzone throughput could significantly reduce the user delay costs, and increase the safety of the high volume highways. This paper involves the evaluation of the workzone's throughput performance. To that end, various strategies such as Multiple Linear Regression, Negative Binomial Regression, and Truncated Regression models were developed to identify the parameters affecting workzone throughputs. An innovative concept of random parameter modeling was also used to account for unobserved heterogeneity issue. Results show that along with traditional factors, further parameters such as number of closed lanes, length of the workzone, presence of police, time of day, and percentage of heavy trucks have a significant effect on the throughput of workzones on high volume highways.