In the petroleum industry, equipments must be maintained properly to meet the adequate reliability standards in order to achieve the desired business goals in terms of productivity, safety and environmental protection. This article offers a new approach focused on risk analysis to select a better maintenance strategy. The proposed approach consists of three stages. In the first step, we identify the accident scenarios that could lead to the loss of production and damage to the environment. In the second step, we estimate the frequency of occurrence of these scenarios. In the third step, we calculate the economic losses and environmental taxes. Finally, an appropriate maintenance strategy is proposed, taking into account the evaluation results obtained by the previous steps. A case study illustrates the proposed approach and shows that the latter constitutes an important decision support tool to improve the existing maintenance strategy to comply with regulations and standards in term of productivity, reduction of costs and environmental protection.

The increasing complexity of most industrial processes always tends to create problems in monitoring and supervision systems. Detection and early fault diagnosis are the best way to manage and solve these problems. Artificial neural networks (ANNs), by their ability to learn and store a large volume of information, are tools particularly suitable for diagnostic support systems. Effectiveness of ANNs for fault diagnosis in milk pasteurization process is presented in this paper. The initial data base used for fault diagnosis is constructed using data extracted from FMEA (Failure Modes and Effects Analysis) tables of milk pasteurization process. Indeed, this analysis makes it possible to establish the links of cause and effect between the faulty components and the observed symptoms. Three models of ANNs, namely Feed-Forward Back Propagation (FFBP), Radial Basis Function based Neural Network (RBNN), and Generalized Regression Neural Networks (GRNN) are developed and compared. The determination coefficient (R2 ), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) statistics were used as evaluation criteria of all the models. The comparison results indicate that the performances of GRNN model are better than the FFBP and RBNN models. The same neuronal models can be extended to any technical system by considering appropriate parameters and defects.

Most managers know that process-risk mapping is essential in enterprise design so as to obtain better understanding and management practices. Organizations need an effective and robust process of management that is less sensitive to changes in the business environment. The main purpose of this paper is the integration of process mapping and risk mapping, with a case study applied in an Algerian company in the oil and gas industry.

In this paper, we will combine a statistical analysis of accident cases concerning the Algerian Oil and Gas Industry occurred during the period of 2003 to 2013 to a dynamic analysis based on the use of a numerical dispersion model. This combination lies in the use of the information obtained from the statistical analysis as input data in the Numerical Dispersion Model (NDM). This study may be useful to illustrate what the industry should learn from these accidents and in such a way be more alert to prevent future major accidents