International Journal of Industrial Engineering: Theory, Applications and Practice
Not a member yet
943 research outputs found
Sort by
A NETWORK-BASED METHOD FOR CONSTRUCTING A TECHNOLOGY ROADMAP
No full textTechnology roadmaps are essential for technology management, but current roadmaps that use text mining to build new processes are time-consuming and laborious. In particular, the identification of relationships between different layers of roadmaps mainly relies on experts. To remedy this problem, this study proposes a systematic and objective process for constructing a technology roadmap. First, by integrating the Stanford parser and a co-word analysis method, we develop a modified Stanford parser-based method to select two-word technological terms, which is suitably comprehensive and complete for building a technology roadmap. Second, we innovatively integrate patents, products, and keywords into the same matrix, thus generating an incidence matrix of products and technologies. Based on this matrix, products and technologies can be directly linked to construct a two-layer product technology roadmap without the assistance of experts. We recommend applying a technology lifecycle to estimate the development time of technologies on the technology roadmap timeline, which is an important but often ignored issue. Finally, we provide an example of cloud computing-related products to demonstrate how the proposed technology roadmap approach can provide managers with a competitive technology and product development strategy
WORKER SCHEDULING OF BANBURY PROCESS USING GOAL PROGRAMMING: A TIRE MANUFACTURING COMPANY CAES STUDY
No full textIn today’s competitive business landscape, organizations must streamline their planning and management processes for efficiency and profitability. Effective resource management, especially in the case of human resources, is crucial. Scheduling challenges in human resource management become more complex when considering worker skills. Balancing fairness for workers and the company necessitates the inclusion of additional constraints and variables, making manual problem-solving more time-consuming and complex. This study proposes a Goal Programming Method, a mathematical model that addresses both primary and goal constraints, to optimize solutions and minimize deviation variables. The case study of a real manufacturer’s Banbury process department with 51 workers, 24 tasks, and two shifts has been analyzed. The mathematical model is utilized to allocate tasks, optimize worker skill utilization, and identify positions to close when worker availability falls short
THE IMPACT OF A VENDOR-MANAGED INVENTORY POLICY ON THE CASH-BULLWHIP EFFECT
Get PDFIt is essential to secure a sustainable flow of cash along the supply chain in the modern corporate environment. Due to the growing significance of cash flow, the concept of the cash-flow bullwhip effect has recently drawn academic attention to finding solutions to the cash shortage. The supply chain’s response to the bullwhip effect, known as the cash-flow bullwhip effect, causes cash-flow volatility to be amplified from downstream to upstream. A number of research studies have looked into the sources and effects of the cash-flow bullwhip effect, but none have concentrated on solutions. This study investigated for the first time the impacts of the vendor-managed inventory policy as a tool for mitigating the cash-flow bullwhip effect. The findings show that the vendor-managed inventory policy typically shortens the cash-conversion cycle of each supply chain member and can, therefore, be implemented as a policy for mitigating the cash-flow bullwhip effect
Path Planning Optimization Algorithm of Track Translation Fruit and Vegetable Picking Manipulator
No full textThe uncertainty of the environment has brought some difficulties to the path planning of the robot arm. In order to improve the path planning effect of the fruit and vegetable picking robot arm, an optimization algorithm for the path planning of the fruit and vegetable picking robot arm with track translation is proposed. By improving the convolution layer of a convolutional neural network, increasing the number of convolution layers and reducing the calculation parameters, the target fruit and vegetable position detection results are obtained to reduce the computational complexity. TOF camera and binocular vision technology are used to judge whether there are obstacles around fruits and vegetables, and SLAM positioning technology is introduced to obtain the three-dimensional position coordinates of target fruits and vegetables and different kinds of obstacles around them, which provides complete information for path planning. The discrete guiding points are obtained by the semantic label map of the road area, and the road guiding lines are obtained by fitting the discrete guiding points with B-spline curves. Through the evaluation function, the guideline information is introduced into the dynamic window method to develop the path planning of the track translation fruit and vegetable picking manipulator so as to achieve the goal of path optimization. The test results show that the proposed algorithm can quickly and effectively realize the path planning of the track translation fruit and vegetable picking manipulator
A New Estimation Approach in Machine Learning Regression Model
No full textIn recent years, machine learning has become a frequently used method for statistical estimation. Random forest regression, decision tree regression, support vector regression and polynomial regression are commonly used supervised machine learning methods. The most commonly used loss function in gradient descent during the optimization phase of these methods is the quadratic loss function, which estimates model parameters by minimizing the cost. The selection of an appropriate loss function is crucial for method selection. There are several loss functions in the literature, such as absolute loss, logarithmic loss and squared error loss. In this study, we propose the use of an inverted normal loss function, which is a finite loss function, to gain a new perspective on minimizing cost and measuring performance in machine learning regression problems. We assert that this loss function provides more accurate estimations of cost minimisation as compared to the quadratic loss function, which is an infinite loss function. This article presents a new approach based on the inverted normal loss function for optimization in regression and performance metrics in machine learning. The procedure and its advantages are illustrated using a simulation study
Multi-Objective Optimization for Sustainable Vehicle Routing Problem Under Uncertainty Using The Lagrangian Relaxation Algorithm-Case: Food Industry Company
No full textThe current study indicates a multi-objective optimization model for vehicle routing problems in the sustainable supply chain under uncertainty conditions. The proposed optimization model seeks to take into consideration the economic, environmental, and social aspects. The focus research is on social indicators among the dimensions of sustainability, for which the weights of the significance of the adverse social effects, including the risk of accidents, working leaves, and the positive social effects, including impartiality between employees, more job opportunities, and heightened levels of welfare for employees, are calculated using the Group Best-Worst method (GBWM) and simultaneously included in the model. Also, the possibilistic-robust programming (PRP) approach was employed to adjust the robustness level of the outputting decisions against the uncertainty of the parameters. A single-objective model can be created by utilizing an extended ε-constraint method from a multi-objective one. and Lagrangian relaxation heuristic is used to solve the proposed model given its medium to large scale, and a case study of a food industry company is examined to verify the applicability of the proposed model for real-life data. The numerical results and the obtained optimal routes indicate that the model can greatly enhance the decision-making capacities of supply chain executives
Maximum Product Spacing Estimation for Odd Lindley Half Logistic Distribution Under Progressive Type-II Censoring with Binomial Removals
No full textIn this paper, we focus on parameter estimation for the odd Lindley half-logistic (OLiHL) distribution under a progressive type II censoring scheme based on binomial removals. The two estimation methods of maximum likelihood and maximum product spacing are employed for the one-parameter OLiHL distribution. To assess the performances of these two estimation methods under different progressive type II censoring schemes formed with binomial removals, a Monte Carlo simulation study is conducted. Moreover, to evaluate the estimation methods under progressive type II censoring, two real data applications are presented from the engineering and medical fields. Based on the results of the simulation study and real data applications, the maximum product spacing method performs better than the maximum likelihood method in estimating the parameter of the OLiHL distribution under progressive type II censoring
Performance Evaluation of Performance Evaluation of Emergency Medical Service Systems with Multiple Ambulance Types and Patient Types
No full textEmergency medical services (EMS) are an important part of the modern healthcare system that tries to provide timely medical care and transportation to patients to reduce morbidity and mortality. Performance evaluation of such EMS systems to determine measures such as mean service rates, dispatch probabilities, busy probabilities, and on-scene times is necessary to design effective and efficient systems. In this paper, we consider an urban EMS system that employs three types of emergency vehicles, including advanced life support (ALS), basic life support (BLS) and first responder vehicle (FRV). We consider two types of patients: type A requires ALS to be dispatched, while type B patients are expected to be served by BLS ambulances. We also consider co-located servers so ambulances of different types can be co-located at the same station. The presence of different types of servers (ambulances) and the patients with different dispatch policies, along with co-located servers, makes it applicable to a more realistic system. We first discuss a modification of the hypercube queueing model for the proposed system and then present an approximate approach for application in large EMS systems. These approaches are compared against a simulation-based model by computing server utilization, service times and on-scene time of ambulances
Analysis of Human Factor Risks in Offshore Platforms Using Human Factors-Failure Mode and Effect Analysis (HF-FMEA)
No full textMany occupational accidents that beset the oil and gas industry globally have been attributed to human errors and other operational constraints. Accident analyses revealed that offshore facilities have a diversified set of equipment focusing on the process requirement rather than system design. This paper aims to create a framework to assess the human factors, hazards or risks associated with maintenance activities within an offshore facility. In pursuit of this goal, a methodology was employed that integrated Failure Mode and Effect Analysis (FMEA) with Human Factors (HF) criteria, resulting in the development of a comprehensive accident analysis framework known as HF-FMEA. This approach considers potential failure modes and also incorporates human factors considerations, thereby enhancing the overall robustness of the analysis. The HF-FMEA framework not only identifies and evaluates failure modes and their potential effects but also takes into account human elements, providing a more holistic understanding of the factors contributing to accidents
A Hybrid Deep Learning based Automatic Target Detection and Recognition of Military Vehicles in Synthetic Aperture Radar Images
No full textATR SAR Imagery is the major application in detection and recognition of military vehicles such as armored vehicles, tank, bulldozer, cannon etc. A robust method that employs Markov Random Field and hybrid Googlenet and VGGnet Convolutional Neural Networks (CNNs) have been proposed in this paper. The performance of Synthetic Aperture Radar (SAR) images is degraded by speckle noise and hence in the first module, we performed SAR image despeckling in order to reduce speckle noise in the images using the improved Adaptive Morphological filter. After despeckling, in the second module, the military vehicles or targets are detected from the despeckled images by Markov Random Field segmentation algorithm. Finally in the third module, hybrid Googlenet and VGGnet Convolutional Neural network with SVM classifier is adopted for classifying and recognizing the military targets from the SAR images. The proposed markov random field with the hybrid VGGnet and Googlenet (VGG-GoogleNet) pretrained Convolutional Neural Networks notably improves the recognition accuracy compared with the conventional deep learning CNN based methods