The modeling approach must related to traffic accident. In addition to the above please forward also the most probable modeling approach and wisely used model in your experience.
Hello Mr. Jima,
As stated, you doubt is complex to answer in a reliable manner. First of all, you need to have the model very clearly defined. What aspect of traffic accidents are you modelling? What is its occurrence rate (broadly speaking)? Where do you get your data from? Is it reliable? After you have all these things clear you need to perform statistical analysis to find out what are its behaviour across time.
Sincerely,
Andre Luiz
You're welcome! Detailing it a bit more and going inwards of the concept involved just remember the any simulation model deals with a set of queues interconnected with its "actions" performed between two (or more) of them. Thus, theoretically speaking queue analysis is yet another aspect that pops up in case you may be interested to expand and go further (even more in traffic because it has a complex queue behaviour).
Cheng & Washington (2005) recommended that the optimum crash history for crash modeling is 3 years and up to 6 years keeping in mind that only minor changes occurred in the study area during this period. [Cheng, W., & Washington, S. P. Experimental evaluation of hotspot identification methods. Accident Analysis & Prevention, 37(5), 870-881, (2005). ]
Negative Binomial regression models are the current state-of-practice for the crash modeling according to the Highway Safety Manual.
Hope that this will help.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling