In this work a new data-driven compression approach is presented. The compression algorithm is an autoencoder trained with an improved On-line sequential Extreme Learning Machine (OS-ELM). First, a dynamic adaptation of the training algorithm towards the newly coming data is achieved by integrating an updated selection strategy (USS) and dynamic forgetting function (DDF). Second, Singular Value Decomposition (SVD) is involved to enhance hidden layer representation via sparse mapping. This new developed autoencoder (ASOS- AE) is compared with the ordinary OS-ELM autoencoder (OS-AE) and proved its accuracy in CMAPSS dataset (Commercial Modular Aero-Propulsion System Simulation). The C-MAPSS software has revisions in civil and military applications. In the present work we used the military version of its applications.

This paper presents a state-of-the-art of recent research work analyzing the requirements of Industry 4.0, particularly related to the competences issue. Over the last few years, the fourth industrial revolution has attracted researchers worldwide to find suitable solutions. However, there are still many gaps related to the Industry 4.0, particularly related to the humans competences issue. Among the many challenges facing companies in this paradigm, one of the most important is the qualification of employees with the necessary skills to succeed in a transformed work environment. To cope with knowledge and competence challenges related to new technologies and processes of Industry 4.0, new strategic approaches for holistic human resource management are needed in manufacturing companies. The main objective of the presented research is to investigate the importance of employee competences, key to the development of Industry 4.0

The era of Industry 4.0 has already begun, however, several improvements should be achieved concerning this revolution. Data mining is one of the modest and efficient tools. Based on a specific query entered in Scopus, related to Industry 4.0, data mining (DM) and logistics, selected documents were studied and analyzed. A brief background of Industry 4.0 and DM are presented. A generic analysis showed that the attentiveness for the cited subject area by countries, universities, authors and especially companies and manufacturers increased through the years. Content analysis reveals that the improvement in quality of the technologies used in manufacturing was noticed, concluding that DM would give Industry 4.0 a leap forward, yet research is dealing with several challenges.

Despite technological advancements, chemical and process industries are still prone to accidents due to their complexity and hazardous installations. These accidents lead to significant losses that represent economic losses and most importantly human losses. Risk management is one of the appropriate tools to guarantee the safe operations of these plants. Risk analysis is an important part of risk management, it consists of different methods such as Fault tree, Bow-tie, and Bayesian network. The latter has been widely applied for risk analysis purposes due to its flexible and dynamic structure. Bayesian networks approaches have shown a significant increase in their application as shown by in the publication in this field. This paper summarizes the result of a literature review performed on Bayesian network approaches adopted to conduct risk assessments, safety and risk analyses. Different application domains are analysed (i.e. accident modelling, maintenance area, fault diagnosis) in chemical and process industries from the year 2006 to 2018. Furthermore, the advantages of different types of Bayesian networks are presented.

In this paper, a new length changeable extreme learning machine is proposed. The aim of the proposed method is to improve the learning performances of a Single hidden layer feedforward neural network (SLFN) under rich dynamic imbalanced data. Particle Swarm Optimization (PSO) is involved for hyper-parameters tuning and updating during incremental learning. The algorithm is evaluated using a subset from C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) dataset of gas turbine engine and compared to its derivatives. The results prove that the new algorithm has a better learning attitude. The toolbox that contains the developed algorithms of this comparative study is publicly available.

Nowadays, the integration of solar power into the electrical grids is vital to increase energy efficiency and profitability. Effective usage of the instable solar production of photovoltaic (PV) systems necessitates trustworthy forecasting information. Actually, this addition can gives an ameliorated service quality if the solar radiation variation can be forecasted accurately. In this paper, we propose a new forecasting approach that integrates Autoregressive Moving Average (ARMA) and Genetic algorithms (GA) to make benefit of both of them in order to forecast Global Horizontal Irradiance (GHI) component. The proposed approach is compared with the standard ARMA model. The experimental results show that, the proposed approach outperforms the classical ARMA models in terms of mean absolute percentage error (MAPE), root mean squared error (RMSE) coefficient of determination (R)2 and the normalized mean squared error (NMSE).

In this paper we analyze a problem of inventory management in a multi-levels multi-products distribution network with three echelon, the studied system consists of a central warehouse and three distribution centers identified by their location zones where each center is connected to a wholesaler group that serve the retailers of his region, which in turn feeds the customers of the regions located in the Algerian territory. The aim of this study is to apply a collaboration between the different actors of the same level in a form of an exchange of products, the exchange can occurs only when the actual demand is being received, in order to study the impact of product exchanges in the distribution networks and its influence on the total costs of the logistics chain from the central warehouse to the delivery to the final customer.

In this work, we introduce the objective of minimizing energy consumption in a job shop scheduling problem with unidirectional transport constraint. In this problem, it is planned to process a set of N jobs (parts) on four machines. The Movement of jobs between these machines is in a single direction that is mean all the parts follow the same direction of movement. Indeed, the energy consumption in this type of problem depends; on the one hand on the speed of the machines processing the jobs and on the other hand on the speed of the means of transport. To solve this optimization problem, we have proposed a metaheuristic method that allows us to find a better sequencing of jobs in order to minimize the cost generated by energy consumption. Several simulations have been studied and the results obtained demonstrate the effectiveness of the proposed approach.

This paper investigates the use of the Particle Swarm Optimization (PSO) algorithm to quantify the effect of RUL uncertainty on predictive maintenance planning. The prediction of RUL is influenced by many sources of uncertainty, and it is required to quantify their combined impact by incorporating the RUL uncertainty in the optimization process to minimize the total maintenance cost. In this work, predictive maintenance of a multi-functional single machine problem is adopted to study the impact of RUL uncertainty on maintenance planning. Therefore, the PSO algorithm is integrated with a random sampling-based strategy to select a sequence that performs better for different values of RUL associated with different jobs. Through a numerical example, results show the importance of optimizing maintenance actions under the consideration of RUL randomness.

In this work a new data-driven approach for Remaining Useful Life estimation of aircraft engines is developed. The proposed approach is a regularized Single Hidden Layer Feedforward Neural network (SLFN) with incremental constructive enhancements. The training rules of this algorithm are inspired form different Extreme Learning Machine (ELM) variants. Particle Swarm Optimization (PSO) algorithm is integrated to enhance tracking ability of the best regularization parameter to reduce the norm of the tuned weights. The proposed approach is evaluated using C-MAPSS (Commercial Modular Aero-Propulsion System Simulation) dataset and compared to its other derivatives and proved its accuracy. C-MAPSS software has revisions in military and civil applications. In this paper, the military version of its application is the used one.

The main objective of this works is to study and improve the performances of the Single hidden Layer Feedforward Neural network (SLFN) for the application of Remaining Useful Life (RUL) prediction of aircraft engines. The most common problems in SLFNs based old training algorithms such as backpropagation are time consuming, over-fitting and the appropriate network architecture identification. In this paper a new incremental constructive learning algorithm based on Extreme Learning Machine algorithm is proposed for founding the appropriate architecture of a neural network under less computational costs. The aim of the proposed training approach is to study its maximum capabilities during RUL prediction by reducing over-fitting and human intervention. The performances of the proposed approach which are evaluated on C-MAPPS dataset and compared with its original variant from the literature. Experimental results proved that the new algorithm outperforms the old one in many metrics evaluations.