Journal of Applied Research on Industrial Engineering

Abstract In today's competitive manufacturing environment, reducing setup times is critical for improving operational efficiency. This case study focuses on applying some of the Seven Tools of Quality (Cause-and-Effect diagram, Individuals and Moving Range control charts, and Pareto chart) using MinitabTM v22 and the PDCA (Plan-Do-Check-Act) cycle to reduce setup times in an automated packaging operation during a five-month study at a corrugated box plant located in the USA Southeast. The pre-intervention (n = 138) began by establishing a baseline average setup time using Individuals and Moving Range control charts, followed by documenting delays and their causes using Pareto analysis. During the "Plan" phase, a brainstorming session with the machine crew and management led to the identification of key areas for improvement. In the "Do" phase, these intervention suggestions were implemented, and post-intervention results were tracked over two months. Post-intervention analysis (n = 153) using the "Check" phase revealed a 23.3% reduction in average individual setup times (from 31.4 min to 24.07 min) and a 64.3% reduction in average moving range setup times (from 17.01 min to 6.08 min). The data points passed the tests for significance using the Western Electric rules. Finally, in the "Act" phase, improvements were reviewed and standardized. The findings highlight the value of combining Lean principles, the PDCA cycle, and the Seven Tools of Quality to achieve significant process improvements without additional capital investment.

Abstract This study explores the comparative efficacy of parametric and nonparametric statistical tests in analyzing clinical metrics, specifically weight (Peso (Kg)), height (H), and Body Mass Index (BMI (kg/m²)), for Glucagon-like Peptide-1 (GLP1) and Sodium-Glucose Co-Transporter-2 (SGLT2) treatment groups. By employing a comprehensive set of statistical methods—including the Sign Test, Wilcoxon Signed-Ranks Test, Wilcoxon Rank-Sum Test, Kruskal-Wallis Test, and Chi-Square Test for nonparametric analysis, and the Paired t-test, Independent Samples t-test, and ANOVA for parametric analysis—we assessed the sensitivity and reliability of these approaches under varying data conditions. Python software was used to ensure accurate and reproducible analyses. Results indicated that parametric tests demonstrated higher sensitivity in detecting significant treatment differences for weight (p = 0.044) and BMI (p = 0.007). Nonparametric methods, although generally more conservative, also revealed significant differences for BMI, including the Wilcoxon Signed-Ranks Test (p = 0.008) and Kruskal-Wallis Test (p = 0.001), highlighting robustness even in conditions of data skewness or non-normality. No significant differences were observed in height measurements across any tests, confirming the specificity of treatment effects to modifiable clinical metrics like weight and BMI. These findings underscore the critical importance of selecting appropriate statistical methods based on data characteristics to enhance accuracy and reliability in clinical research.

Abstract The COVID-19 pandemic led to significant disruptions across global industries, with far-reaching economic effects that continue to shape operational practices worldwide. Driven by the need to understand sector-specific weaknesses and international variations better, this study investigates the impact of the pandemic on industrial sectors in ten countries heavily affected by the crisis: China, France, Germany, Iran, Italy, South Korea, Spain, Switzerland, the United Kingdom, and the United States. Employing input–output analysis alongside scenario modeling, the study explores seven potential outcomes based on the early-stage disruptions observed during the pandemic. The analysis focuses on key industries, including manufacturing, transportation, and services. The findings reveal that Italy faced substantial losses in industrial output, ranging from 7.24% under the best scenario to 8.04% under the worst. In comparison, China experienced more modest declines, from 1.90% to 2.49%. Sectors including transportation, hospitality, and retail were particularly hard hit, with Iran experiencing the most severe losses across industries. These results provide critical insights into sectoral resilience and emphasize the need for targeted risk management approaches and coordinated policies to aid recovery and bolster economic stability in the event of future global crises.

Abstract Ventures are currently collaborating and forming supply chain relationships under specific strategies, such as a vendor-managed inventory (VMI) system. In a VMI system, the vendor manages all inventory decisions instead of the buyer, enhancing supply chain clarity, reducing costs, and potentially improving its environmental sustainability. Our proposed research examines a two-level supply chain with a single vendor and a single buyer handling constantly deteriorating chemical products, such as ethylenediamine, diethyl oxalate, sodium borohydride, and zirconium tetrachloride, whose demand follows a time-dependent quadratic form. These products require proper preservation to minimize deterioration while also addressing the impact of carbon emissions from transportation as a contributor to climate change. Three cases are discussed for the VMI and non-VMI systems:(i) green investment to reduce carbon emissions; (ii) preservation technology to reduce deterioration rates; and (iii) a combination of preservation technology and green investment to reduce both deterioration and carbon emissions. The objective is to minimize the total supply chain cost per unit time, considering the decision variables, and the classical optimization method used to derive the optimum solution of objective functions. An inventory cost analysis, both with and without VMI, demonstrates that the VMI system in our proposed model consistently reduces overall supply chain costs and outperforms the conventional approach. The findings indicate that the VMI system, when combined with preservation and green investments, incurs the lowest cost compared to other cases. Numerical examples and a sensitivity analysis are provided to validate the derived results.

Abstract Reducing defects in manufacturing presents a critical issue in achieving cost-effective, high-quality production. Though machine learning (ML) and Bayesian optimization (BO) have been independently tested against defect prediction and process optimization, their combined use has not been extensively studied. This study seeks to fill the potential gap in the literature by suggesting a hybrid framework that integrates ML models with a BO mechanism to predict the likelihood of production defects and determine the optimal production parameter sets, consequently minimizing the possible defects. The preprocessing of the dataset of key manufacturing variables included using ADASYN to deal with data imbalances in classes and the Las Vegas Filter (LVF) to select important features. Optuna was employed to train and fine-tune nine ML models, with Random Forest indicating the best performance (accuracy: 94.6%). Bayesian optimization was subsequently utilized to deduce the best values of ten critical features and reduce defect probability to the minimum possible (up to 0.000925). The results supported the effectiveness of the suggested method in predicting with high precision and actionable optimization, offering a scalable solution consistent with the Industry 4.0 objectives.

Abstract The purpose of this paper is to develop a new Failure Mode and Effect Analysis (FMEA) framework for identification, prioritization and improvement of failure modes. A hybrid multiple criteria decision-making (MCDM) method combining Stepwise Weight Assessment Ratio Analysis (SWARA) and VlseKriterijumska optimizacija I KOm-promisno Resenje (VIKOR) is used to rank the risk of failure modes identified in FMEA. For this purpose, the SWARA method is utilized to obtain the influential weights and then VIKOR technique is employed to give the prioritization levels for the failure modes in safety equipment. A case study of a gas company in Lorestan Province (Iran) is provided to illustrate the potential application and benefits of the proposed FMEA approach. The obtained results show that the new risk priority model can be effective to find high risk failure modes and create suitable maintenance strategies. The proposed FMEA also can overcome the shortcomings, improve the effectiveness of the traditional FMEA and provide useful information to help in managing risks of safety equipment.

Abstract Predicting the results of sports matches is interesting to many, from fans to punters. It is also interesting as a research problem, in part due to its difficulty, because the result of a sports match is dependent on many factors, such as the morale of a team (or a player), skills, current score, etc. So even for sports experts, it is very hard to predict the exact results of sports matches. This research discusses using a machine learning approach, Artificial Neural Networks (ANNs), to predict the outcomes of one week, specifically applied to the Iran Pro League (IPL) 2013-2014 football matches. The data obtained from the past matches in the seven last leagues are used to make better predictions for the future matches. Results showed that neural networks have a remarkable ability to predict the results of football match results.

Abstract The main purpose of this paper is to identify the traditional, green and effective resilience criteria in the performance of green and resilient suppliers and their ranking with path analysis, SWARA and TOPSIS combined approach in Fanavaran Petrochemical Company. The research method is applied in terms of goal and descriptive-survey in terms of data collection. By a comprehensive review of the literature, first a set of key performance criteria and sub-criteria (traditional, green, and resilience) were extracted. Then, using the path analysis approach, the effectiveness of these criteria was evaluated in Fanavaran Petrochemical Company. The statistical population included 55 experts of the mentioned company, which due to the limited size of the population, all members were considered as the research sample. The path analysis result showed that all identified criteria affect the company’s supplier’s performance. Then, using new SWARA decision-making technique and also the opinions of 30 experts, the criteria and sub-criteria were evaluated and their weight (importance) was extracted. In the final evaluation of the main criteria, the criterion of “resilience” was in the first rank, the criterion of “green” in the second rank and the criterion of “traditional” in the last rank. Subsequently, due to the sensitivity of the ranking of green and resilient suppliers in the company, using the TOPSIS decision-making technique and based on the extractive weight of the criteria, seven suppliers of the company were evaluated by the experts and the final ranking of the suppliers in terms of performance was determined. Thus, the proposed approach of this research provides a valuable conceptual framework for company’ managers to improve the situation of the suppliers in terms of the environmental issues and resilience. Also, the development and improvement of traditional criteria and selection of suppliers of the company based on green standards and resilience were the main goals of this research.

Abstract This paper, presents a mathematical model of a four-level supply chain under hyperchaos circumstances. The analysis of this model shows that the hyper-chaotic supply chain has an unstable equilibrium point. Using Lyapunov's theory of stability, the problem of designing a hyperchaotic supply chain control is investigated. The design of the nonlinear controller is performed first to synchronize two identical hyper-chaotic systems with different initial conditions and then to eliminate the chaotic behavior in the supply chain and move to one of unstable equilibrium points, as well as different desired values ​​at different times. A different supply chain is predicted to demonstrate the performance of the controller. In the next part of numerical simulation, with the control of the distributor as the center of gravity of the model, the stability of the entire chaotic supply chain can be achieved.  The most important point in designing a control strategy is the ability to implement it in the real world. Numerical simulation results in all stages show that the applied nonlinear control policy can provide supply chain stability in a short period of time, also, the behavior of control signals has low amplitude and oscillations. In other words, it represents a low cost to control the hyperchaotic supply chain network.

Abstract 5S is an important industrial engineering technique which is used worldwide in a wide range of industrial and service type organizations for workplace management. Improvement in efficiency and productivity, and reduction in waste and idle time etc. are some of its benefits. This paper presents a fundamental understanding of 5S technique and review of some important past work on implementation of 5S in various organizational setups. It is worth mentioning that safety has been identified as to be included as the 6^thS under this technique. The main aim of this paper is to facilitate scholars, researchers, and engineers of industrial engineering field by providing knowledge and develop understanding of 5S technique so that they may further implement it in various scenarios of the workplace organization.

Abstract The recognition of pathways and identification of cars was seen with a prospective camera, which recognizes trajectories and predicts control points. The aim is to propose the location of the path. In this paper, lane detection algorithm Steering Assistance System (SAS) is introduced. Guiding helps to learn driving and anticipates the control points and defines the direction that makes it easy to learn in a potential way and a lane keeping assistance system which warns the driver on unintended lane departures. Path keeping is an important element for self-driving cars. This article describes the beginning to end adapting the approach to holding the car in the right direction.

Abstract The evolution of the means aimed at improving the availability of strategic equipment requires a credible level of maintenance with the development of original monitoring techniques such as vibration diagnostics. The vibration monitoring of strategic equipment through a vibration diagnosis requires mainly the identification of vibrational images of the different types of damage. However, one of the important vibration problems is essentially due to the phenomenon of imbalance. This phenomenon corresponds to an imbalance of the rotor due to the offset between the axis of inertia and the axis of rotation, which causes significant and cyclical vibrations. The aim of this study is to analyze the vibratory behavior of a screw compressor to improve its reliability and consequently its availability. To identify the imbalance fault, two sets of vibration measurements on April and January were fundamentally examined at the compressor level. Fourier transform based on spectral analysis was used to create a vibration detection approach with vibration signals. The comparison of the results obtained with that of the simulation resulting from the model of dynamic behavior of the compressor is conclusive.

Abstract Scheduling is an important decision-making process that aims to allocate limited resources to the jobs in a production process. Among scheduling problems, Hybrid Flow Shop (HFS) scheduling has good adaptability with most real world applications including innumerable cases of uncertainty of parameters that would influence jobs processing when the schedule is executed. Thus a suitable scheduling model should take parameters uncertainty into account. The present study develops a multi-objective Robust Mixed-Integer Linear Programming (RMILP) model to accommodate the problem with the real-world conditions in which due date and processing time are assumed uncertain. The developed model is able to assign a set of jobs to available machines in order to obtain the best trade-off between two objectives including total tardiness and makespan under uncertain parameters. Fuzzy Goal Programming (FGP) is applied to solve this multi objective problem. Finally, to study and validate the efficiency of the developed RMILP model, some instances of different size are generated and solved using CPLEX solver of GAMS software under different uncertainty levels. Experimental results show that the developed model can find a solution to show the least modifications against uncertainty in processing time and due date in an HFS problem. 

Abstract In this paper, a one-dimensional fractional advection-diffusion equation is considered. First, we propose a numerical approximation of the Riemann-Liouville fractional derivative which is fourth-order accurate, then a numerical method for the fractional advection-diffusion equation using a high order finite difference scheme is presented. It is proved that the scheme is convergent. The stability analysis of numerical solutions is also discussed. The method is applied in several examples and the accuracy of the method is tested in terms of error norm. Furthermore, the numerical results have been compared with some other methods.