International Journal of Industrial Engineering: Theory, Applications and Practice
Not a member yet
943 research outputs found
Sort by
AN ALTERNATIVE DESIGN FOR THE VARIABLE SAMPLE SIZE COEFFICIENT OF VARIATION CHART BASED ON THE MEDIAN RUN LENGTH AND EXPECTED MEDIAN RUN LENGTH
No full textControl charts for monitoring the coefficient of variation (CV) are attracting increasing attention in recent years. These charts are able to monitor processes with an unstable mean and/or standard deviation, but has a stable CV. Existing CV charts are designed based on the average run length (ARL) criterion. However, this paper has shown that designs based on the ARL criterion could result in misinterpretation of the chart’s actual performance. Hence, this paper proposes alternative designs for the VSS CV chart, based on the median run length (MRL) and expected median run length (EMRL) criteria. This paper also compares the performance of the VSS CV chart with that of the Exponentially Weighted Moving Average (EWMA) and Shewhart CV charts, based on the proposed designs. Subsequently, this paper shows the implementation of the proposed designs on an industrial example.
Spreadsheet Error Detection and Debugging Approach for Dynamic Discrete Inventory Control Models
No full textSpreadsheets have been widely used in many business and scientific areas for modeling and simulation. Nevertheless, a number of research surveys have shown that these applications are particularly prone to errors. Some of the reasons for this error-proneness can be attributed to the fact that spreadsheet models are usually developed by the end-users and that standard software quality assurance processes are typically not applied. In addition, end-users often don't have a "clear picture" about the model structure and elements represented in the spreadsheet. The paper presents error detection and debugging approach developed for dynamic discrete spreadsheet models of inventory control. Spreadsheet modeling of inventory dynamics is based on the utilization of the discrete-time system control concept. This concept provides additional rules (common and specific constraints) which can be used for error detection. Preliminary experiments show general applicability of the approach
AN ANALYSIS OF NONLINEAR INVENTORY-PRODUCTION CONTROL SYSTEM WITH PRODUCTION CONSTRAINTS
No full textIn case of free fluctuation production process, some portion of the capacity (including equipment and human resource) would be underutilized so that production managers in factories almost produce in a smooth manner with less deviation from average. In order to model this smoothed production strategy, Inventory and Order Based Production Control System (IOBPCS) is extended by introducing upper and lower constraints on production process to restrict highly fluctuated manufacturing resulting in more smooth operation with less cost compared with the swing production. Several linear and nonlinear extensions have been made on IOBPCS family but there is no production smoothing constraints on it. The extended model is more realistic and is called it NIOBPCS (Nonlinear IOBPCS). NIOBPCS has more advantages compared with IOBPCS in terms of and filtering out market demand noises, bullwhip effect reduction, less production fluctuation, and high capacity utilization
Automatic Guided Vehicle Systems in Flexible Manufacturing System –A review
No full textMaterial handling operations in flexible manufacturing systems has become efficient and easy with the improvement in automated machine technology. Material handling has been improved immensely since it started as fully manual operations, where men were employed to lift, stack and count the jobs in any manufacturing facility. Today’s rapid development in technology presents manufacturing firms a range of alternatives for in-plant transportation. Automated Guided Vehicle system (AGVs) are applied for advance material handling practices in a Flexible manufacturing facility consisting of a number of autonomous vehicles (AGVs) for movement of work part from one work center to another by following a particular guide path and their operations are controlled by a computer program. Now a day’s agent based AGV are also used in Material handling operations which operates and perform their assigned task autonomously. In this paper AGV system design and control in manufacturing facilities along with state of art review of the literature available on AGV guide path configurations, comparison of guide path configurations, estimation of the number of AGVs in a facility, scheduling and dispatching policies, routing, conflict resolution and positioning of AGVs has been reviewed and at the end of the paper, literature on Integrated policies for optimum utilization of AGVs is reported
A KERNEL FISHER DISCRIMINANT ANALYSIS-BASED TREE ENSEMBLE CLASSIFIER: KFDA FOREST
No full textIn general, an ensemble classifier is more accurate than a single classifier. In this study, we propose an ensemble classifier called kernel Fisher discriminant analysis forest (KFDA Forest). This is a tree-based ensemble method that applies KFDA. To promote diversity, a bootstrap is used and variable sets are randomly divided into K subsets. KFDA is performed on each subset to increase classification accuracy. KFDA maximizes the distance between classes while minimizing the distance within classes. KFDA can also be applied to classification problems in a nonlinear data structure using the kernel trick because it can transform the input space into a kernel feature space, commonly named rotation, rather than a dimensionality reduction. Because new feature axes and KFDA projections are parallel, decision trees are used as a base classifier. To compare the proposed method with existing ensemble methods, we apply these to real datasets from the UCI and KEEL repositories
MULTIMODAL OPTIMIZATION OF JOB-SHOP SCHEDULING PROBLEMS USING A CLUSTERING-GENETIC ALGORITHM BASED APPROACH
No full textThough ample work has been done in optimization of job shop scheduling problems (JSSPs), very few techniques can satisfy the requirement of modern day workshop i.e. providing multiple schedules that achieve the desired goal. These techniques also do not provide a good balance between converging to the optimal solution while finding multiple optimal solutions. To overcome this obstacle, a new algorithm is proposed, by combining k-means clustering algorithm and Genetic Algorithm (GA), for the multimodal optimization of JSSPs. In the proposed algorithm, k-means clustering algorithm is first utilized to cluster the individuals of every generation into different clusters, based on some machine sequence related features under the assumption that different global optima will have different features. Next the adapted genetic operators are applied to the individuals belonging within the same cluster with the aim of searching for global optima within each cluster independently. The performance of the proposed algorithm is measured by its application to the multimodal optimization of benchmark JSSPs and comparing its performance against other multimodal optimization algorithms. The results of the case studies show that the algorithm has better average optimal value and is also capable of finding multiple optimal solutions
AN IOT DATA ANOMALY RESPONSE MODEL FOR SMART FACTORY PERFORMANCE MEASUREMENT
No full textThe recent rapid proliferation of Internet of Things increases the possibility of IoT failure due to complexity of the network and large data volumes. The demand for clear acquisition of data as an important source of performance measurement presents manufacturing companies with a major challenge. To cope with IoT fault problem, we developed the performance measurement and IoT anomaly response model. Development comprises three steps: (1) configuration of the IoT-based performance measurement process using an Overall Equipment Effectiveness Key Performance Indictor and IoT anomaly response model architecture; (2) implementation of an IoT fault case classification and data anomaly detection and mitigation algorithm, using K-means and statistical inference methods; and (3) validation of the proposed algorithm through experimental simulation. The experimental simulation results show that the proposed algorithm has a positive impact on IoT data anomaly detection and mitigation, thus enabling a response to the IoT fault problem
SCHEDULING APPOINTMENTS OF TRUCK ARRIVALS AT CONTAINER TERMINALS
No full textIn container terminals at hub ports, each trucking company delivers a large number of containers every day. The truck drivers for the delivery operation may experience long waiting times when they arrive at peak hours. To reduce congestion, this study assumes that the container terminal charges a congestion fee (appointment charge), which depends on the expected staying time of trucks in the terminal, to control truck arrival times. This study proposes a scheduling method for appointments that takes into account the appointment charge, waiting time, operation cost of trucks, and number of available trucks. This study introduces a mathematical formulation and heuristic algorithms for solving the problem within a reasonable computational time. The heuristic algorithms consist of Phase 1, which solves a transportation problem, and Phase 2, which generates a solution respective of that obtained via Phase 1 such that the capacity constraint of the trucks is satisfied. Numerical experiments are conducted to test the heuristic approaches
A Dynamic Failure Rate Prediction Method for Chemical Process System under Insufficient Historical Data
No full textBecause of large number of equipments, long pipelines and complex process, the chemical process system is especially prone to accidents due to component defects and equipment failures, making it tough challenging to guarantee process security. According to the relationship between process state parameters and fault conditions in chemical process, and taking into account that in some circumstance the historical fault data is insufficient to perform statistical analysis, this paper proposes a method to predict the dynamic failure rate of chemical process system based on BP (back-propagation) neural network and two parameter Weibull distribution. First, the BP neural network is applied to expand the limited amount of data of process state parameters and determine the fault states. Then combining the expanding data set, some mathematical methods are applied to determine the parameters of Weibull distribution and the failure rate function, based on which the mean failure rate can be calculated for each phase. A liquid chlorine storage system of a chemical plant is introduced to demonstrate the proposed method. Compared with the traditional method and the method that merely considers known limited amount of fault data into two-parameter Weibull distribution, the results show that the failure rate of the liquid chlorine storage system calculated by the proposed method is more consistent with the actual situation, especially more closer to the actual failure rate of the system in the early stage. Moreover, based on the expansion of historical data, this method can achieve a continuous dynamic prediction for future failure rate and failure time points, which has practical meaning for the prevision and dissolving of accident risk in chemical plants
EPQ models with production rate proportional to power demand rate, rework process and scrapped items
No full textIn the most of the production systems, at the end of the replenishment period, some imperfect units detected by inspection process, are reworked and the others are disposed. Constant rate of demand is one of the usual assumptions in the conventional lot sizing models. But this assumption is not suitable in many real cases. With regard to these points, this work deals with a joint problem of demand having a power pattern, production rate pro rata with demand rate, defective items, rework process and scrap items. It is assumed that a manufacturer may be faced with two scenarios: all imperfect units are recovered or a certain fraction of imperfect units are reworked and the others are disposed. Setup, holding, backordering, inspection, production, rework and disposal costs are involved in the inventory system. An algorithm is presented to optimize total inventory cost and determine the best reorder point, lot size and scheduling period. A numerical analysis is carried out to illustrate the applications of the proposed models.