Publications Scientifiques

Permanent URI for this communityhttps://dspace.univ-boumerdes.dz/handle/123456789/10

Browse

Filters

2011 - 2019172020 - 202419

Settings

Author: search.filters.author.Mellal, Mohamed Arezki

Search Results

Now showing 1 - 10 of 36
  • No Thumbnail Available
    Item
    Multi-objective optimization of series-parallel system with mixed subsystems failure dependencies using NSGA-II and MOHH
    (SAGE, 2024) Mellal, Mohamed Arezki
    In complex systems, failure dependencies play a crucial role in determining their overall performance. This paper explores the multi-objective optimization of series-parallel systems with mixed failure dependencies. By optimizing system cost and availability, the study aims to identify the most efficient redundancy and repair strategies. Two optimization algorithms, the non-dominated sorting genetic algorithm II (NSGA-II) and a novel multi-objective algorithm named the multi-objective hoopoe heuristic (MOHH), are utilized alongside constraint handling techniques to produce Pareto fronts. These fronts illustrate the trade-offs between cost and availability. Additionally, a fuzzy decision method is utilized to determine the best compromise solutions from each optimization technique. Comparing the results, NSGA-II consistently outperforms MOHH in providing better compromise solutions across five independent runs. However, MOHH demonstrates a better standard deviation in its performance.
  • No Thumbnail Available
    Item
    A novel approach for remaining useful life prediction of high-reliability equipment based on long short-term memory and multi-head self-attention mechanism
    (Wiley-Blackwell, 2024) Al-Dahidi, Sameer; Rashed, Mohammad; Abu-Shams, Mohammad; Mellal, Mohamed Arezki; Alrbai, Mohammad; Ramadan, Saleem; Zio, Enrico
    Accurate prediction of the Remaining Useful Life (RUL) of components and systems is crucial for avoiding an unscheduled shutdown of production by planning maintenance interventions effectively in advance. For high-reliability equipment, few complete-run-to-failure trajectories may be available in practice. This constitutes a technical challenge for data-driven techniques for estimating the RUL. This paper proposes a novel data-driven approach for fault prognostics using the Long-Short Term Memory (LSTM) model combined with the Multi-Head Self-Attention (MHSA) mechanism. The former is applied to the input signals, whereas the latter is used to extract features from the LSTM hidden states, benefiting from the information from all hidden states rather than utilizing that of the final hidden state only. The proposed approach is characterized by its capability to recognize long-term dependencies while extracting features in both global and local contexts. This enables the approach to provide accurate RUL estimates in various stages of the equipment's life. The proposed approach is applied to an artificial case study simulated to mimic the realistic degradation behaviour of a heterogeneous fleet of aluminium electrolytic capacitors used in the automotive industry (under variable operating and environmental conditions). Results indicate that the proposed approach can provide accurate RUL estimates for high-reliability equipment compared to four benchmark models from the literature.
  • No Thumbnail Available
    Item
    Improvement of system reliability in a natural gas processing facility by PSO and DE
    (Springer Nature, 2024) Saheb, Tafsouthe; Mellal, Mohamed Arezki
    The reliability of the systems as well as its optimization is the first concern of the designers. The elements of a given system can be either in series, parallel, parallel-series, or in a complex configuration. This paper addresses the reliability optimization of a natural gas processing facility. The reliability of this system is calculated and two redundancies strategies, active and standby, are optimized under the resource limits to improve reliability. Two bio-inspired optimization algorithms, namely the particle swarm optimization (PSO) and the differential evolution (DE), are implemented with penalty functions to find the optimal redundancy. The results obtained are compared.
  • No Thumbnail Available
    Item
    Combined heat and power economic dispatch problem with binary method using flower pollination algorithm and differential evolution
    (Springer, 2023) Mellal, Mohamed Arezki; Khitous, Marwa; Zemmouri, Meriem
    Nowadays, the need for electrical energy became crucial in the world. The co-generation plants, which simultaneously produce electrical and heat energies, are one of the alternative solutions to supply people and industry with both energies. The present work addresses the cost minimization of the nonconvex combined heat and power dispatch problem (CHPED). The nonconvex operating region is handled using the binary method, and the optimization problem is solved using two nature-inspired algorithms, namely the flower pollination algorithm (FPA) and the differential evolution (DE). Penalty functions are adopted to handle all the operating constraints, units’ limits, and demands. The results obtained compare the algorithms and those of the literature. It is observed that the fuel cost obtained by the flower pollination algorithm (FPA) is less than the one obtained by the differential evolution (DE) and the particle swarm optimization (PSO)
  • No Thumbnail Available
    Item
    Optimal reliability allocation of heterogeneous components in pharmaceutical production plant
    (Springer, 2023) Aqel, Ibrahim; Mellal, Mohamed Arezki
    he COVID-19 pandemic and competitiveness pressure the pharmaceutical companies to acquire systems designed to be as reliable as possible. The present paper aims to optimize the design of a pharmaceutical plant through the reliability allocation of heterogeneous components under the design constraints. The problem is solved by resorting to three nature-inspired algorithms of artificial intelligence (AI): grey wolf optimizer (GWO), shuffled frog-leaping algorithm (SFLA), and adaptive particle swarm optimization (ADAP-PSO). A penalty function is implemented to handle the constraints and the results obtained are compared
  • No Thumbnail Available
    Item
    System design optimization with mixed subsystems failure dependencies
    (Elsevier, 2023) Mellal, Mohamed Arezki; Zio, Enrico; Al-Dahidi, Sameer; Masuyama, Naoki; Nojima, Yusuke
    Systems present dependencies among their components failure behavior, which impact their ultimate availability. Previous works addressed the optimal design of systems in relation to its cost and under given availability constraint, considering identical subsystems failure dependencies. The present paper addresses this problem in a realistic scenario by taking into consideration mixed subsystems failure dependencies. The problem is formulated with reference to a complex bridge network system and a series-parallel system. Three nature-inspired optimization techniques are implemented to solve the problem, namely differential evolution (DE), manta ray foraging optimization (MRFO), and shuffled frog leaping algorithm (SFLA) with constraint handling. A numerical evaluation is performed; the results show that DE outperforms MRFO and SFLA
  • No Thumbnail Available
    Item
    Resource allocation modeling framework to refactor software design smells
    (2023) Gupta, Priyanka; Anand, Adarsh; Mellal, Mohamed Arezki
    The domain to study design flaws in the software environment has created enough opportunity for the researchers. These design flaws i.e., code smells, were seen hindering the quality aspects of the software in many ways. Once detected, the segment of the software which was found to be infected with such a flaw has to be passed through some refactoring steps in order to remove it. To know about their working phenomenon in a better way, authors have innovatively talked about the smell detection mechanism using the NHPP modeling framework. Further the authors have also chosen to investigate about the amount of resources/efforts which should be allotted to various code smell categories. The authors have developed an optimization problem for the said purpose which is being validated on the real-life smell data set belonging to an open-source software system. The obtained results are in acceptable range and are justifying the applicability of the model
  • No Thumbnail Available
    Item
    Guest editorial : reliability and quality : analysis and applications
    (Emerald Group Holdings, 2022) Bhargava, Cherry; Sharma, Pardeep Kumar; Patil, Rajkumar Bhimgonda; Mellal, Mohamed Arezki
  • No Thumbnail Available
    Item
    Reliability, availability, maintainability, and dependability analysis of Tube-wells Integrated with Underground Pipelines in agricultural fields for irrigation
    (2022) Kumar, Ashish; Saini, Monika; Rajkumar Bhimgonda, Patil; Sameer, Al-Dahidi; Mellal, Mohamed Arezki
    Reliability, Availability, Maintainability, and Dependability (RAMD) study of Tube-wells Integrated with Underground Pipelines (TIUP) is crucial as they are the backbone of the irrigation system. This study is carried out with an objective to perform RAMD analysis, and Failure Modes and Effects Analysis (FMEA) unified with the development of a novel stochastic model using Markovian approach to estimate the Steady-State Availability (SSA) of the TIUP. A real case study of a conventional TIUP system has been performed to validate theoretical and practical results of the proposed model. The failure and repair rates of all subsystems followed exponential distribution, and their impact on system/subsystem’s availability and other reliability measures has been investigated. All the repairs are perfect and random variables associated with failure and repair rates are statistically independent. The centrifugal pump and power supply units are the most critical components as far as reliability and maintainability aspects. The labor also plays a critical role in the operation of the TIUP system
  • No Thumbnail Available
    Item
    Multi-objective factors optimization in fused deposition modelling with particle swarm optimization and differential evolution
    (Springer, 2022) Mellal, Mohamed Arezki; Laifaoui, Chahinaze; Ghezal, Fahima; Williams, Edward J.
    The design of any system contemplates the elaboration of a prototype of the entire system or some parts, before the manufacturing phase. Nowadays, rapid prototyping (RP) is widely used by the designers. Achieving good manufacturing performances needs to handle various process parameters. Most works deal with single objective process parameters. The reality is quite different and the processes involve conflicting objectives. This paper addresses the multi-objective factors optimization of the fused deposition modelling (FDM) technology. The problem is converted into a single one using the weighted-sum method and then solved by resorting to two nature-inspired computing techniques, namely particle swarm optimization (PSO) and differential evolution (DE). The results obtained are compared