CONTACT US: Contact info
|
Citation |
Seise A, Peltola H, Leden L, Virkkunen M. Res. Rep. Finn. Transp. Agency 2012; (38): online. |
|
Copyright |
(Copyright © 2012, Finnish Transport Agency) |
|
DOI |
unavailable |
PMID |
unavailable |
|
Abstract |
Since 1999, a systematic inventory has been carried out of conditions and traffic volumes at Finnish level crossings between state-owned railways and roads. We compiled accident data from these level crossings and studied the effects of various variables on accident numbers. Accident prediction models were created to predict overall numbers of level crossing accidents and significant accidents (as defined by ERA). The main factors affecting accident numbers were: number of road and rail vehicles using the level crossing, warning devices, speed limit on the road and rail and sight conditions after bush removal, and type of road surface. Using the collected data, a new safety evaluation tool was created. A similar tool called Tarva (for estimation of traffic safety effects of improvements using impact coefficients) had previously been created for Finnish and Lithuanian highways. The newly created program, Tarva LC, made it possible to (i) review the factors affecting level crossing safety, (ii) evaluate the current safety of all the level crossings on the state rail network as reliably as possible, (iii) estimate the safety effects of level crossing improvements, and (iv) study the cost-effectiveness of such improvements. Combining data from accident prediction models and accident history, it is possible to estimate the expected number of accidents at each level crossing if no improvements are implemented. To estimate the safety effects or benefits of making improvements, we need both the expected number of accidents without those improvements (estimated as mentioned above) and the results of research from other countries showing how much accident numbers are reduced from such measures being implemented. Knowing the average implementation costs, one can then calculate the cost-effectiveness of various measures or combinations of measures. Having the expected numbers of accidents available and using common evaluation principles as well as same effect coefficients not only help in making evaluations but also enhance the comparability of results. |