The main results delivered in the project can be categorised under two main impact areas. First, the project aimed to improve the safety data collection and analysis, for instance by harmonising taxonomies and methods, selecting best-in-class standards to improve the analysis by «digging deeper». Second, best practices need to be disseminated and adopted in the community by various stakeholders, ranging from ship-plane manufacturers, operators of various size, service providers, regulators and national agencies. SAFEMODE applied user-centred principles to all of its activities, with frequent interactions and feedback from stakeholders, to define sustainable methods, proportionate to the potential level of risk, coherent with the organisational safety and Human Factors culture.
"Improving safety data collection and analysis" area
The main results are the following:
- A unified Taxonomy between aviation and maritime to analyse safety occurrences, as a means to improve Safety Learning. The SHIELD taxonomy has been tried and tested on hundreds of aviation and maritime incidents and accident reports, to ensure it is usable and helpful in understanding the event and all its contributory factors. Following this application, the initial version of SHIELD was simplified to strike the right balance: "as simple as possible, but not too simple".
- Design and development of the SHIELD Database to collect data and identify top risks. SHIELD is both a taxonomy and a database. The SHIELD database gives more power to the analysis, because it can go beyond single events, or even a company’s own fleet experience, to consider a broader range of safety events from which learning can be extracted. The SHIELD database allows designers to see what went right and what went wrong with similar or related designs.
- Test of advanced analytics to mine safety data and extract insights. SAFEMODE tested and compared the performance of various Natural Language Processing algorithms on SHIELD and on other freely available safety datasets, e.g. NASA ASRS. For each algorithm, we assessed its performance and limits of applicability, for instance the minimum number of occurrences needed for the algorithm to perform adequately.
- Development of the HF COMPASS tool, to guide users into picking the best HF method & technique. The Human Factors COMPASS addresses one of main challenges reported by stakeholders, i.e. being unsure about which Human Factors methods they should apply for their projects and in which order.
- Guidance for applying, building and quantifying Risk Models for aviation and maritime. Risk models determine which barriers and elements are vulnerable, primarily human actions with influences. This risk model is applicable to a wide range of situations and typically supports: i) Identifying high-risk areas, ii) identifying and assessing the impact of new design solutions, iii) supporting incident investigations.
"Spreading best practices" area
The main results are:
- Application of the HURID framework to 7 different Case Studies, involving different partner organisations, with different levels of HF expertise and experience. In particular, the Wake Vortex Alert Case Study was performed to provide a state-of-the-art example of how HF methods can steer and inform design decisions, from initial co-design workshop to human-in-the-loop simulations on the cockpit and ground side. An innovative Low Fidelity Simulation approach was tested for innovative concepts development. On the maritime side, Case Studies addressed research on Human Response to Emergencies with human-in-the-loop simulation of bridge operations, forward looking scenarios of remotely piloted unmanned ships, better integration of Human Factors into a shipping company Safety Management System.
- Two output papers were presented to the International Maritime Organisation, on the SHIELD-HF taxonomy and Safety Learning Culture. The first submission is entitled “Proposal for a new output to amend the Casualty Investigation Code to mandate a root cause investigation” and was rejected by Flag states. The second submission is entitled "Report on safety learning culture". The proposal was accepted and the main Safety Committee invited the Member states to submit papers based on the SAFEMODE outcome presented. This includes even new output proposals to start a new agenda item to take further action relating to the concept of a safety learning culture. Finally, the IMO Maritime Safety Division scheduled a SAFEMODE presentation at the IMO (5 June 2023, 12:45) on Safety Learning Culture during the next MSC 107.
- Authoritative online resources like SEAbrary - an electronic repository of maritime safety knowledge related to maritime operations, management, and design with a focus on Human Factors (HF) - and SafeFlix - video introductions and/or tutorials about selected Human Factors methods.
- Training and workshops were delivered in EU and international (Philippines, Indonesia) countries, reaching approx. 500 participants.
