Network-wide road crash risk screening: A new framework.

Network-wide road crash risk screening is a crucial issue for road safety authorities in governing the impact of road infrastructures over road safety worldwide. Specifically, screening methods, which also enable a proactive approach (i.e., pinpointing critical segments before crashes occur), would be extremely beneficial. Existing literature provided valuable insights on road network screening and crash prediction models. However, no research tried to quantify the risk of crash on the road network by considering its main components together (i.e., probability, vulnerability, and exposure). This study covers this gap by a new framework. It integrates road safety factors, prediction models and a risk-based method, and returns the risk value on each road segment as a function of the probability of a crash occurrence and the related severity as well as the exposure model. Next, road segments are ranked according to the risk value and classified by a five-level scale, to show the parts of road network with the highest crash risk. Experiments show the capability of this framework by integrating base map data, context information, road traffic data and five years of real-world crash data records of the whole non-urban road network of the Province of Brescia (Lombardy Region - Italy). This framework introduces a valid support for road safety authorities to help identify the most critical road segments on the network, prioritise interventions and, possibly, improve the safety performance. Finally, this framework can be incorporated in any safety managerial system.

Prediction of the severity of exceeding design traffic loads on highway bridges.

Being a driver of failure consequences, forecasting the severity of events where design traffic load limits on bridges have been exceeded (DLEEs) is fundamental for road safety. Previous research has focused on estimating failure consequences by direct and indirect cost metrics. Only recently has research assessed severity unconventionally, in which the type of DLEEs was predicted by applying econometric models through Binomial Logistic Regression (BLR). Since machine learning models using Artificial Neural Networks (ANN) have not yet been explored, this study will enhance the literature as follows. First, two different 'severity' models were set up as a function of bridge-side, temporal-context, and traffic load hazard variables. Whilst the former relied on a BLR, the latter used an ANN. Second, the performance of these models was assessed using confusion matrixes, some performance indicators, and a cross-entropy parameter. Raw Weigh-In-Motion data on 7.4 M+ individual vehicle transits on a bridge along a primary roadway in Brescia (Italy) were processed. Although a similarly strong performance was achieved for BLR and ANN, the results indicated that ANN was able to predict severity records with a higher level of confidence than BLR on the case study dataset, with the cross-entropy of the ANN less than one third of that of the BLR. These analyses can support road authority traffic management to safeguard bridges from traffic load hazards. Finally, this study recommends future developments, such as considering the structural effects of traffic loads in the modelling, prioritizing traffic management actions among bridges at network level, and exploring the impact of ANN models in risk assessment.

Bus crash risk evaluation: An adjusted framework and its application in a real network.

Greater attention to bus safety can lead to relevant benefits for public transport companies in terms of higher service performance, reliability, and lower insurance costs. Therefore, measuring the crash risk on bus routes provides an opportunity to improve the safety performance of transit operators. Previous research has explored the effects of many factors regarding the frequency and severity of bus crashes, whereas only a handful of studies have defined some crash risk indexes. Conversely, to the best of our knowledge, almost no research has been done regarding the crash risk in the bus transit network that integrates frequency, severity, and the exposure factors. This paper proposes a new framework to assess the crash risk for each transit bus route by the integration of safety factors, prediction models and risk methods. More precisely, this framework identifies several safety factors and specifies the risk components in terms of frequency, severity and exposure factors that may affect bus crashes. Then, it models their relationships to build a bus crash risk function. Lastly, according to the values returned by the previous function, the crash risk for each route is computed and a safety performance ranking for each route is provided. The feasibility of this framework is demonstrated in a real case study by using bus crash data provided by a mid-sized Italian bus operator. The findings show that transit managers could implement this framework in a road traffic safety management system to evaluate the risk of crashes on routes, monitor the safety performance of each route and qualify each route according to recent safety norms.