Road traffic crash is a problem in every country. According to World Health Organization, in Europe there are about 127,000 people killed and at least some 2.4 million injured each year. Incident detection algorithms to detect incidents on motorways are well developed. Their objective is to detect incidents as soon as possible so that emergency services can get to the scene of the incident as soon as possible to reduce congestion, to provide medical help if needed and to increase safety of the affected area. Some crashes are traffic related and it may be possible to detect the state of the traffic where the risk of a crash is high. Based on this, premise safety indicators have been developed by [Aron et al, 2003], [Hayward, J. C., 1972], [Lee, C. et al, 2006], and [Pande, A. et al, 2006]. The derivation of these safety indicators are data driven and may be able to forecast the potential of a crash in real time. This study aims to assess the driving risks under Swiss road conditions by using safety indicators. Real traffic data from Automatic Traffic Counts installed on motorways and crash data have been collected in Vaud canton, Switzerland. This paper also discusses the applicability of the proposed safety indicators and the issues associated with their use.