64
6 Results and research
contributions
65
explained by company-specific circumstances (e.g. management involvement, resources for dealing with safety issues, and safety training of employees).
6.1.2 Paper II
Jacobsson, A., Ek, Å., Akselsson, R. (2011). Method for evaluating learning from incidents using the idea of “level of learning”. Accepted for publication in Journal of Loss Prevention in the Process Industries.
This paper has the same background and overall objective as Paper I, but the topic is the learning product – the lesson learned. A method is described for evaluating the effectiveness of learning, based on the level of learning. The level of learning is expressed in terms of how broadly the lesson learned is applied geographically, how much organisational learning is involved and how long-lasting the effects, according to a classification system. To evaluate the actual and the potential levels of learning and comparing the two, a 6-step method was developed.
1. Evaluation of the actual level of learning, based on the lessons learned from individual reported incidents.
2. Evaluation of the potential level of learning from individual reported incidents.
3. The relationship between the actual and potential levels of learning for a larger number of incidents.
4. Adjusting the results from 1-3, taking into consideration incidents that are not reported (the hidden number).
5. Consideration of possible learning from incidents on an aggregated basis.
6. Consideration of other learning mechanisms related to incidents.
Tools were developed to help in making concrete numerical evaluations. A key step in the method is the evaluation of the full causation picture with all its underlying causes which enable the user of the method to draw conclusions regarding the potential learning from the incidents. The method contains a specific tool for this.
The great value of the method is not the generated numbers per se but the message they convey when the numbers are related to the level of learning they stand for and when comparisons over time or between companies or departments are made. The method can give clear indications of areas for improvement. The measures of the method can also be used in efforts to correlate results from learning from incidents with other safety parameters (e.g. results from safety audits and safety climate inquiries). The method is intended to be used on a sample of the broad range of incidents normally seen in process industry companies.
The method was tested on incident reports covering two years from the six companies in the LINS study from various types of the process industry. It was found that the method and the tools developed worked very well in practice.
66
The results varied substantially between the companies. However, on average it can be said that 25% of the incidents resulted in level 0 learning (No learning), 50% in level I learning (Limited local level), 18% in level II learning (Local level), 6% in level III learning (Process unit level) and 1% in level IV learning (Site level). The ratio actual/potential learning varied between 0.36 and 0.86 (without adjustment for the hidden number).
Similar to Paper I, the results provided insights into what can influence the effectiveness of learning from incidents. Differences between the companies could often be explained by company-specific circumstances (e.g. management involvement, resources for dealing with safety issues and safety training of employees).
6.1.3 Paper III
Jacobsson, A., Sales, J., Mushtaq, F. (2009). A sequential method to identify underlying causes from industrial accidents reported to the MARS database. Journal of Loss Prevention in the Process Industries 22 (2), 197-203.
This paper presents a method designed to identify underlying causes leading to industrial accidents. The method is generic in nature but closely associated with the way the reporting to the MARS database of the European Commission denominates and structures the causes of accidents. The method developed intends to facilitate the learning process from accidents by identifying possible causes related to the accidents that were not directly stated in an accident report, but that can be deduced following the description of the event. This is particularly the case with regard to the components and quality of the safety management systems in place at the industrial establishment at the time of the accident. The method follows a sequential approach, although a combination of the philosophy behind other existing accident models has been taken into consideration. The starting point of the model is the causes for accidents included in the MARS database. These causes have been extended by considering typical operational or organisational failures that are normally related to the original reported cause(s). The extension of causes has been performed by adding three follow-on levels of possible underlying causes. The first level can be considered as a direct cause of the accident and, the last level being more applicable to the foundation of establishing safety: “Safety Management System or the Safety Culture”.
The objective is to determine the effectiveness of the method in identifying underlying causes in addition to those causes stated in the original reports. In this way, it is possible to establish a system to go deeper into the analysis of past accidents, in order to obtain lessons learned, and to avoid the recurrence of similar accidental scenarios in the future, as well as to give directions for a better reporting system of industrial accidents.
67
The method was applied to the total set of accidents reported to the MARS database.
It was found that the method was easy to use and it is argued in the paper that the causation model developed is suitable for its purpose, which was to expand the causation analysis of accidents to include more underlying causes. The method also received great support from a group of experts of the European Federation of Chemical Engineering.
The main results of the analyses are that as much as three times as many underlying causes can be found when applying the method developed compared with what is given in the original reports.
The method in this paper was also developed specifically to be used for analysing the accidents of the MARS database to find possible characteristic patterns of the underlying causes, and the potential for learning from the accidents, which should be a reflection of the underlying causes.
6.1.4 Paper IV
Jacobsson, A., Sales, J., Mushtaq, F. (2010). Underlying causes and level of learning from accidents reported to the MARS database. Journal of Loss Prevention in the Process Industries 23 (1), 39-45.
One of the main purposes of the MARS database is to provide information for learning from the accidents to avoid similar events. The main objective of this paper was therefore to determine how good the learning from the accidents reported to MARS actually is. Other objectives were to establish whether there were any specific patterns per industry type and per country in the learning. A specific objective was to establish whether there had been any impact of the requirement in the Seveso II legislation regarding safety management system on the causes of accidents.
Two separate measures were used as indicators of the learning:
1. the extent to which relevant causes have been analysed 2. the level of learning of the lesson learned
It is argued that the most important issue for the learning from accidents is the analysis of the causes of the accident, particularly the underlying causes, which are the key to deciding on relevant measures. The sequential method, presented in Paper III, made it possible to go beyond the causes given in the original reports and to find more underlying causes. A classification system was developed to determine the level of learning from the accidents using the actions/lessons learned given in the reports.
This method establishes the level of learning of the lessons learned from each case description, essentially from the breadth of application and from an organisational point of view.
68
The paper presents results from an analysis of all the accidents reported to the MARS system up to mid-2007 regarding the underlying causes and the extent of learning, based on the level of learning.
Both the methods used, the one for analysis of underlying causes and the one for establishing the level of learning, worked very well on the data in the MARS database.
The most important underlying causes were found in weaknesses in process analysis (risk assessment) and in procedures, regardless of industry type. Weaknesses in safety management systems and in safety culture contribute as underlying causes in a very high percentage of the accidents. No major differences in the pattern of the underlying causes were found for the various industry types, neither for the various countries. The quality of reporting, measured in terms of analysis of underlying causes, vary considerably between various countries. The level of learning, as determined from the information in the reports, is found to be in general rather low, especially from some of the countries. In two thirds of the accidents the learning stops at a local level within the sites. This study resulted in ideas of improvement of the MARS system.