Overall safety performance of Air Traffic Management system: Forecasting and monitoring Giulio Di Gravio a , Maurizio Mancini b , Riccardo Patriarca a , Francesco Costantino a,⇑ a Department of Mechanical and Aerospace Engineering, University of Rome ‘‘La Sapienza’’, Italy b ENAV S.p.A, Via Salaria 716, 00138 Rome, Italy article info Article history: Received 23 June 2014 Received in revised form 2 October 2014 Accepted 6 October 2014 Available online 11 November 2014 Keywords: ATM safety Aerospace Performance Factor Prevention Monte Carlo Time Series analysis Mixture model abstract Defining means to assess safety performance and delve into their causes is one of the current and future challenges of the air transport sector. This paper develops an improvement of the Air Traffic Management (ATM) safety evaluation in order to develop proactive safety indicators, based on Aerospace Performance Factor and Analytic Hierarchy Process. The research aims to carry out a statistical model of safety events in order to predict safety performance, combining in a Monte Carlo simulation the results emerged from the literature analysis with the analytical models of historic data interpretation. Through the analysis of the possible scenarios, assessing their impact on equipment, procedures and human factor, this model will address the interventions of the decision maker. Ó 2014 Elsevier Ltd. All rights reserved. 1. Introduction Based on international regulations, ANSPs must plan a rugged and proactive process of addressing current and emerging safety risks, in order to ensure that air traffic development is carefully managed and supported through strategic regulatory and infrastructure development (ICAO, 2013). Historically, ANSPs used basic metrics as traffic counts, number of accidents and incidents to gauge safety performances. Anyway, these standard indicators fail to represent effectively the overall safety perspective and do not constitute a system-wide performance measurement tool. In October 2009, the EUROCONTROL Performance Review Commission (PRC) and the US Federal Aviation Administration (FAA) identified common information and performance indicators (EUROCONTROL and FAA, 2012) to monitor safety in each region. EUROCONTROL Safety Regulatory Requirements (ESARRs) proposed a standard occurrence reporting and assessment scheme. In partic- ular, ESARR 2 Appendix A (and B) (EUROCONTROL, 2009) provides the minimum contextual/factual ATM related (and no-ATM related) to be collected and recorded for each safety occurrence. The core idea of ESARRs is based upon Reason Swiss Cheese model, which relates a system’s failure to an alignment of all the metaphoric barriers’ weakness, permitting ‘‘a trajectory of accident opportunity’’, so that a hazard passes through all of the holes in all of the defences, leading to a failure (Reason, 1990; Stranks, 2007). Fig. 1 represents this core concept, according to a possible safety event. In ATM context, for example, even when many things go wrong in case of a separation infringement between aircrafts, i.e. the situ- ation is not recognized or resolved by ATC, pilots or TCAS, the air- craft still only have a small chance to hit each other, due to the fact that there is a lot of space in the sky. In other words, there will be many near midair collisions compared to the number of midair collisions, although the differences of these events may only be coincidence (the aircraft colliding in midair had exactly the wrong position and velocity with respect to each other). The difference between such a near midair collision and a midair collision might be therefore just be the lining up of the small hole in the last geometry slice of cheese. Note that other factors may cause the difference between a near collision and a collision, e.g. pilot adequately performs a good TCAS instruction. Therefore, it is evident that the evaluation of global safety performance cannot disregard the contribution of any safety related event, especially the ones with smaller consequences. As less serious events happen more often, statistics based on their occur- rences have more potential than accident statistics. Investigations of occurrences less serious than accidents might also be available more quickly. This concept leads to the metaphor of an iceberg where the most serious occurrences -accidents and serious http://dx.doi.org/10.1016/j.ssci.2014.10.003 0925-7535/Ó 2014 Elsevier Ltd. All rights reserved. ⇑ Corresponding author at: Via Eudossiana 18, 00184 Rome, Italy. Tel.: +39 06 44585260; fax: +39 06 44585250. E-mail address: francesco.costantino@uniroma1.it (F. Costantino). Safety Science 72 (2015) 351–362 Contents lists available at ScienceDirect Safety Science journal homepage: www.elsevier.com/locate/ssci