Preface to the Special Issue “Mathematical Modeling in Industrial Engineering and Electrical Engineering”—Special Issue Book

It is now clear that cooperation between academia and industries is crucial for social, cultural, technological and economic progress and innovation. Cooperation between these realities is an essential tool for the development of knowledge, guaranteeing greater competitiveness. The conditions for a virtuous exchange of knowledge between universities and industries are set in motion by cooperating to find ways, languages and opportunities to achieve the necessary coordination: requests and offers from industries and the supply of knowledge and skills from universities.

Thus, in the university world in general, and in particular in the fields of electrical and industrial engineering, the need arises to carry out research projects shared with the world of industry. There is no doubt that mathematical modeling is the first important step in the management of industrial problems. However, such models are often complex and require numerical techniques to obtain solutions, especially when the huge amount of input data is affected by uncertainty.

This Special Issue aims to explore, from a broad perspective, the most recent developments in the field of mathematical modeling for problems of interest in electrical and industrial engineering. The response of the scientific community has been significant, with many papers being submitted for consideration, and, finally, twenty-one papers were accepted, after going through a careful peer-review process based on quality and novelty criteria.

The paper authored by Y.E. Shao and S.C. Lin [1] presents an innovative time-delay neural network-based classifier to diagnose the quality variables that cause out-of-control signals for a normal multivariate process with variance shifts. To demonstrate the effectiveness of the approach, simulated experiments were conducted whose results were compared with artificial neural network classifiers, support vector machine and multivariate adaptive regression spline classifiers finding that the proposed classifier accurately recognizes the contributors of the out-of-control signal.

The paper by J. Park [2] presents an efficient pipeline transmission algorithm with the techniques of changing the transmission order between nodes considering the communication status of the processing nodes (changing the transmission order for the transmission operation based on the communication status of the processing). Upon receipt of a transmission operation, a local bus checks the size of the remaining pre-existing transmission data of each processing node by transmitting the data in the modified transmission order.

The paper authored by P. Di Barba, L. Fattorusso and M. Versaci [3] presents a new second-order nonlinear differential model with stationary singularity for membrane electrostatic 2D MEMS devices in which the amplitude of the electric field is locally proportional to the curvature of the membrane The paper presents an important result of the existence of at least one solution.

The paper by G. Angiulli, S. Calcagno, D. De Carlo, R. Laganà and M. Versaci [4] proposes a new dynamic model based on physical transmission of heat by conduction to analyze and predict the thermal stresses in the plate of a commercial aircraft. The model has provided thermal stress maps completely similar to those that can be experimentally obtained using infrared thermal imaging cameras, highlighting the evolution of the thermal load of the steel plate attack wing-fuselage, adding evidence of possible phenomena of input fatigue to identify in advance if the steel plate needs to be replaced.

The paper by A. Jannelli [5] deals with the numerical solutions of a class of fractional mathematical models that arise in the engineering sciences governed by time-fraction of advection-diffusion-reaction equations, involving the Caputo derivative (especially in chemical and hydrodynamic processes) proposing an implicit finite difference methodology unconditionally stable. The results show that the proposed procedure is efficient, reliable and easy to implement.

The paper authored by K.M. Kim, S.-H. Choe, J.M. Ryu and H. Choi [6] exploits the Padé approximation to solve the zoom locus problem by structuring and analytical form of a rational function constituted by the ratio of polynomials. The results obtained open a new frontier for solving this type of problem.

The paper authored by W. Liu, S. Song, Y. Qiao and H. Zhao [7] studies the coordination of the supply chain where the retailer is loss averse and a combine d buyback and quantity flexibility agreement is introduced. In the paper, which has as its primary aim the maximization of the conditional Value-at-Risk of utility, it is shown that the combined contract coordinates the chain and there is a single coordinating wholesale price if the confidence level is below a threshold. In addition, the retailer’s optimal order quantity, expected utility, and coordinated wholesale price are decreasing in terms of loss aversion and confidence levels, respectively.

The paper by OP.C.J. Neto, E.G. Gonzalo, F.S. Lasheras and A.B. Sanchez [8] presents an innovative machine learning procedure exploiting the hybrid gradient enhanced regression tree (GBRT) to predict hard chromium layer thickness using a prediction model consisting of a weak prediction model set. Here, GBRT hyper-parameters are optimized by differential evolution. The proposed model is able to predict the chromium layer with a high performance allowing the classification of the importance of the input variables.

The paper authored by C.-N. Wang, T.-T. Dang and N.-A.-T. Nguyen [9] develops innovative computational models for inventory control in Thailand, focusing on parameters such as order quantity, reorder point, stock target and inventory review techniques, with the primary aim of determining the best levels of factors significantly affected by their responses. The results obtained confirm that the selected judgment parameters are correct and performing, providing a useful guideline for decision-making choices to satisfy customer demand at the minimum total inventory cost possible.

The paper authored by W. Liu, S. Song, Y. Quiao, H.Zhao and H. Wang [10] studies a reference-dependent loss-averse newsvendor problem (with stochastic demand and rate of return) by obtaining the loss-averse newsvendor optimal sorting policy. In the work, it is proved that the optimal order quantity of the loss averse newsvendor and expected utility decrease in loss aversion level and benchmark. Furthermore, even if the effect of the random return leads to an increase in the quantity of the order, the loss averse seller can order more, equal or less than the classic one, which depends significantly on the level of loss aversion and the point of reference.

The paper by Y. Peng and J. Wu [11] studies a stochastic fluid queuing system led by Lévy where the server is subject to failure and/or repair. By exploiting the joint use of the Lévy and Kella-Whitt processes, the limiting distribution of the workload process was achieved, also investigating the period occupied and the correlation structure. Finally, it has been proved that the stochastic decomposition properties also hold for fluid tails with Lévy inout.

The paper authored by M. Versaci and G. Angiulli [12] proposes an alternative and numerically performing approach for the evaluation of the electrostatic capacitance of metal cylinders. The proposed numerical procedure, based on latest generation Krylov subspaces, is able to provide usable results for larger conductors in which the single module can be represented by the metal cylinder studied in the paper.

The paper authored by L.R. Merchan-Villalba, J.M. Lozano-Garcia, J.G. Avina-Cervantes, H.J. Estrada-Garcia, A. Pizano-Martinez and C.A. Carreno-Meneses [13] proposed the design of a decoupled linear control strategy for a dynamic voltage restored (DVR) exploiting matrix converter (MC) as its core element achieving the compensation energy directly from the power system. The designed system is able of addressing the gaps in power quality (including harmonic distortion) establishing an adequate transient response for the converter in terms of convergence speed and overshoot magnitude, while also ensuring the stability of the closed loop system under limited operating conditions. Further measures have made it possible to develop a multifunctional compensator that is easy to manufacture.

The paper by M. El-Borhamy, E.E.M. Rashad, I. Sobhy and M.K. El-Sayed [14] proposes an innovative semi-analytical procedure for the prediction of stable periodic responses of AC electrical machines capturing the correct selections of machine variables highlighting that, compared to the experimental results, the derived analytical results are relatively well suited to the practical cases studied.

The paper by Y.-M. Tu [15] deals with a short-term scheduling model of cluster tool in wafer fabrication maximizing throughput within time constraints estimating the arrival time and planning in the short term. The dynamic cycle time of the product phase was applied to estimate the arrival time of the work in process (WIP) in front of the machine. Without violating the time constraint of the WIP, an algorithm was also developed to calculate the last time of the WIP. In the work, it was found that, on the basis of the last process time of the WIP and the efficiency table of the combination, the production program of the cluster tools can be reorganized to achieve the production aims.

A. Montisci, S. Cargangiu, G. Sias, B. Cannas and A. Fanni, in [16], present a real time bolometer tomographic reconstruction algorithm in nuclear fusion reactors based on the combination of highly correlated grid elements to yield a square coefficient matrix with sufficiently low condition number. The matrix, inverted offline, can be multiplied by the actual bolometric measurements providing real-time tomographic reconstruction.

J. Liu, L. Zhou and Y. Wang, in their work [17], aim to produce an LCECSC consisting of a single electronic platform and a single manufacturer, providing two models with or without altruistic preference. In addition, in this work, the authors combine the characteristics of online sales by assuming that altruistic preference is a proportional function of commission, thus establishing an extended model based on commission with altruistic preference. The results achieved showed that, unlike the traditional offline supply chain, the dominator electronic platform profit is less than the follower producer profit. In addition, as consumers’ carbon reduction elasticity coefficient increases, the service level, selling price, carbon reduction, sales, supply chain member profits, and systems profits increases, ultimately improving economic and environmental performance. Finally, the altruistic preference behavior of the electronic platform is a profit transfer behavior.

The paper authored by J. Wang, P. Gao, Z. Li and Q. Bai [18] has as its main purpose the construction of a model for forecasting the cycle time (CT) unified under dynamic work-in-progress levels. Particularly, the work proposes a transfer learning approach to fine-tune the predicted neural network in a hierarchical way. Starting from the structuring of a 2D convolutional neural network for CT prediction under a primary work-in-progress level with the input of space-time characteristics by rearranging the input parameters, a hierarchical optimization transfer learning strategy was proposed, tested and validated to fine-tune the prediction model to improve the accuracy of the CT prediction.

The paper by P. Li, X. Li, J. Li, Y. You and Z. Sang [19] deals with a real-time harmonic extraction approach for distorted grid in order to extract harmonics quickly and precisely for the sustainability of an electrical system. The proposed extraction method is based on a time-varying observer using a real-time estimation algorithm of the current frequency of the electricity grid. Following the conversion of the frequency into a suitable phase variable, an observer was modeled to extract each harmonic component. The proposed method has been evaluated on a FPGA test platform, highlighting that the proposed procedure is capable of extracting harmonic components of the mains current and converge within 80 ms even in the presence of mains distortions.

In the paper produced by C.-Y. Lee, G.-L. Zhuo [20], an innovative model of failure diagnosis for rolling elements was proposed. IT consists of feature extraction and selection for fault classification where feature extraction was combined with signal processing to reduce the noise. Signal processing was done by local mean decomposition (LMD) reducing signal noise by product function selection (PF) and decomposition wavelet package (WPD) extracting he fault information. The performance of the model has also been improved through an optimization technique (IBPSO) by extracting the most important features. The feature extraction process effectively removes noise and establish a high-precision feature set. In addition, the proposed feature selection algorithm has higher accuracy than the other state-of-the-art feature selection algorithms. Finally, in a highly noisy environment, the performance of the fault diagnosis model was compared with the performance of the state-of-the-art fault diagnosis model highlighting that the proposed technique maintains a high level of classification accuracy.

The paper authored by M. Cotronei and C. Moosmueller [21] proposes a new development of the Hermite polynomial splines through a quick procedure for the calculation of the refinement mask of the Hermite B-spline of any order and in the case of a general scaling factor. The procedure was derived from the polynomial reproduction properties satisfied by the Hermite splines without requiring the explicit construction/evaluation of the basic functions. In addition, the authors discussed in detail the factorization properties of Hermite’s B-spline masks in terms of the increased Taylor operator, proving to be the minimum annihilator for the discrete monomial Hermite sequence space of fixed degree.

As Guest Editor of this Special Issue, I would like to express my gratitude to all the Authors who, for various reasons have contributed with their papers. I also thank all the Reviewers who, with their comments and/or suggestions, have made possible the realization of this Special Issue.

The purpose of this Special Issue was to attract quality and new documents in the field of analytical-numerical modeling for problem of industrial interest and, in particular of electrical engineering problems.

It is hoped that the papers selected and presented in this Special Issue will have an impact on the international scientific community by motivating further research in industrial sectors which, today are expanding.