Mahmoud A. Barghash

Pattern recognition of control charts using artificial neural networks—analyzing the effect of the training parameters

In this paper, we have utilized artificial neural networks (ANN) for pattern recognition of the most common patterns which occur in quality control charts. After detecting such patterns, it is possible to relate these patterns to their causes. This could find extreme importance for on-line quality monitoring and on-line trouble shooting. It could be possible to detect problems before they become serious and the operator has to shut the line down or the process may produce defective parts. In this work, we have attempted to explore the effect of the training parameters on the performance of the neural network. The training parameters are important because they emphasis the required performance and the accuracy required from the neural network. A resolution IV fractional factorial experiment is utilized to explore a portion of the range of selected parameters to obtain better performance of the neural network. The results showed that many parameters usually assigned by experience such as minimum shift, shift range, population size and shift percentage, have significant effect on the performance of the ANN, while others such as network size and window size do not have major significance on the performance of the net.

A generalised framework for simulation-based decision support for manufacturing

Simulation in general and discrete event simulation in particular has been widely used in manufacturing since the 1960s. The increased computational power accompanied by reduction in cost and success of simulation increased the areas of applications of simulation as a decision support tool for manufacturing. However, along with this increased usage of simulation, a formal scheme that organises the use of simulation in manufacturing can be critical for the optimal use of this tool in a routine manner. In this work, a Simulation Application Framework for Manufacturing (SAFM) is built. The framework that highlights the nature of relations between different decision areas, and makes the task of using simulation to support decisions more systematic, is built based on reviewed literature. Main components of the framework are: manufacturing strategies, layout, material handling, scheduling and manufacturing processes and resources. In the SAFM, the use of simulation is triggered and controlled by managerial strategies and priorities.

Shrinkage and Warpage Detailed Analysis and Optimization for the Injection Molding Process Using Multistage Experimental Design

ABSTRACT This work suggests a multistage experimentation for modeling and optimization of the shrinkage and warpage of injection molding parts. This methodology has moderate experimental size while analyzing all important interactions. Phase 1 is Taguchi experimentation, which determines significant factors, and Phase 2 is full factorial. This technique reduced the number of experiments from 729 to 108. It also assisted in modeling shrinkage and warpage up to an error value of 0.23 and 0.07 mm, respectively. It also resulted in better optimal process settings than single-stage Taguchi experimental design while performing fewer experiments than full factorial with all factors included.

Dynamic programming model for multi-stage single-product Kanban-controlled serial production line

The executive concern of this paper is how to control and synchronize the flow of materials in Kanban controlled serial production line so as to build a dynamic material-flow system that successfully meets customer demand Just-In-Time. The proposed approach should yield a consistent integrated control policy with a feasible level of Work-In-Process and a feasible corresponding operational cost. The production line is described as queuing network, and then a Dynamic Programming (DP) algorithm is used to solve the network by decomposing it into several numbers of single-stage sub-production lines. Backward computations of DP are done recursively with synchronization mechanism, in the since that the solution of one sub-production line is used as an input to the previous one. A performance measure is then developed to determine and to compare the values of production parameters. Numerical examples are used to demonstrate the computations of different system parameters, the results are validated by discrete events simulation using ProModel software version 6.0, the performance measure coincided with the results of the model with very small error (0.044). As a result the number of Kanbans that are needed to deliver the batches from upstream stage to the downstream stage is determined in such a way that keeps the stages synchronized with the external customer demand.

An improved hybrid algorithm for the set covering problem

Discussing a number of weak points of a previous algorithm to solve the set covering problem (SCP).Developing a new hybrid algorithm that has the best performance among all meta-heuristics to solve the SCP.Proposing a new mechanism to update the pheromone trails limits in a Max-Min Ant System (MMAS).Using a simple normalizing step to deal with possible ranges of heuristic information.Very low computation times. The state-of-the-art ant colony optimization (ACO) algorithm to solve large scale set covering problems (SCP) starts by solving the Lagrangian dual (LD) problem of the SCP to obtain quasi-optimal dual values. These values are then exploited by the ACO algorithm in the form of heuristic estimates. This article starts by discussing the complexity of this approach where a number of new parameters are introduced to escape local optimums and normalize the heuristic values. To avoid these complexities, we propose a new hybrid algorithm that starts by solving the linear programming (LP) relaxation of the SCP. This solution is used to eliminate unnecessary columns, and to estimate the heuristic information. To generate solutions, we use a Max-Min Ant System (MMAS) algorithm that employs a novel mechanism to update the pheromone trail limits to maintain a predetermined exploration rate. Computational experiments on different sets of benchmark instances prove that our proposed algorithm can be considered the new state-of-the-art meta-heuristic to solve the SCP.

An effective and novel neural network ensemble for shift pattern detection in control charts

Pattern recognition in control charts is critical to make a balance between discovering faults as early as possible and reducing the number of false alarms. This work is devoted to designing a multistage neural network ensemble that achieves this balance which reduces rework and scrape without reducing productivity. The ensemble under focus is composed of a series of neural network stages and a series of decision points. Initially, this work compared using multidecision points and single-decision point on the performance of the ANN which showed that multidecision points are highly preferable to single-decision points. This work also tested the effect of population percentages on the ANN and used this to optimize the ANNs performance. Also this work used optimized and nonoptimized ANNs in an ensemble and proved that using nonoptimized ANN may reduce the performance of the ensemble. The ensemble that used only optimized ANNs has improved performance over individual ANNs and three-sigma level rule. In that respect using the designed ensemble can help in reducing the number of false stops and increasing productivity. It also can be used to discover even small shifts in the mean as early as possible.

An ANN-GA Framework for Optimal Engine Modeling

Internal combustion engines are a main power source for vehicles. Improving the engine power is important which involved optimizing combustion timing and quantity of fuel. Variable valve timing (VVT) can be used in this respect to increase peak torque and power. In this work Artificial Neural Network (ANN) is used to model the effect of the VVT on the power and genetic algorithm (GA) as an optimization technique to find the optimal power setting. The same proposed technique can be used to improve fuel economy or a balanced combination of both fuel and power. Based on the findings of this work, it was noticed that the VVT setting is more important at high speed. It was also noticed that optimal power can be obtained by changing the VVT settings as a function of speed. Also to reduce computational time in obtaining the optimal VVT setting, an ANN was successfully used to model the optimal setting as a function of speed.