Papers by abdelkader mazouz
Journal of Real-Time Image Processing, 2021
Training of convolutional neural networks (CNNs) on embedded platforms to support on-device learn... more Training of convolutional neural networks (CNNs) on embedded platforms to support on-device learning has become essential for the future deployment of CNNs on autonomous systems. In this work, we present an automated CNN training pipeline compilation tool for Xilinx FPGAs. We automatically generate multiple hardware designs from high-level CNN descriptions using a multi-objective optimization algorithm that explores the design space by exploiting CNN parallelism. These designs that trade-off resources for throughput allow users to tailor implementations to their hardware and applications. The training pipeline is generated based on the backpropagation (BP) equations of convolution which highlight an overlap in computation. We translate the overlap into hardware by reusing most of the forward pass (FP) pipeline reducing the resources overhead. The implementation uses a streaming interface that lends itself well to data streams and live feeds instead of static data reads from memory. ...
Proceedings of SPIE, Feb 1, 1992
The reliability of the materials handling process involving automated stacking of packages on a p... more The reliability of the materials handling process involving automated stacking of packages on a pallet or automated sorting of packages in a distribution system depends mainly on the design of the package and the material used for the package. Many problems can be eliminated that result in a higher utilization of the system if the package is designed not only for the product and its requirements but also for an automated handling system with different types of grasping devices. A decision support system is being developed to help the package designer select the most appropriate material and design to satisfy the requirements of the automated materials handling process. The decision support system is programmed in C++ which gives the flexibility and portability needed for this type of system. The user interface is using graphics to ease the understanding of different design options during the selection process.
Proceedings of SPIE, Feb 19, 1988
CRC Press eBooks, Aug 29, 2000
A two-layer perceptron feedforward backpropagation network architecture with a minimum number of ... more A two-layer perceptron feedforward backpropagation network architecture with a minimum number of hidden neurons is designed using the backpropagation training algorithm for a noncomplex application, namely, controlling a plant inventory system. The design objective is to determine the number of hidden neurons and what type of data must be entered to get the backpropagation algorithm started. The convergence of the
Seventh International Conference on Intelligent Systems Design and Applications (ISDA 2007), Oct 1, 2007
Methods of modeling the hydrologic process range from human observers to sophisticated surveys an... more Methods of modeling the hydrologic process range from human observers to sophisticated surveys and statistical analysis of climatic data. In the last few years, researchers have applied computer programs called neural networks or artificial neural networks to a variety of uses ranging from medical to financial. The purpose of the study was to demonstrate that commercial neural networks can be successfully applied to hydrologic modeling. Using publicly available climatic and stream flow data and a ward systems neural network, the study resulted in prediction accuracy of greater than 97% in +/-100 cubic feet per minute range. The average error of the predictions was less than 16 cubic feet per minute. To further validate the model's predictive capability, a multiple regression analysis was done on the same data. The neural network's predictions exceeded those of the multiple regression analysis by significant margins in all measurement criteria.
Ecology and the Environment, May 4, 2007
For the entire period of recorded time, floods have been a major cause of loss of life and proper... more For the entire period of recorded time, floods have been a major cause of loss of life and property. Methods of prediction and mitigation range from human observers to sophisticated surveys and statistical analysis of climatic data. In the last few years, researchers have applied computer programs called Neural Networks or Artificial Neural Networks to a variety of uses ranging from medical to financial. The purpose of the study was to demonstrate that Neural Networks can be successfully applied to flood forecasting. The river system chosen for the research was the Big Thompson River, located in North-central Colorado, United States of America. The Big Thompson River is a snow melt controlled river that runs through a steep, narrow canyon. In 1976, the canyon was the site of a devastating flood that killed 145 people and resulted in millions of dollars of damage. Using publicly available climatic and stream flow data and a Ward Systems Neural Network, the study resulted in prediction accuracy of greater than 97% in +/-100 cubic feet per minute range. The average error of the predictions was less than 16 cubic feet per minute. To further validate the model's predictive capability, a multiple regression analysis was done on the same data. The Neural Network's predictions exceeded those of the multiple regression analysis by significant margins in all measurement criteria.
Energies, Aug 8, 2019
Optimal expansion of medium-voltage power networks is a common issue in electrical distribution p... more Optimal expansion of medium-voltage power networks is a common issue in electrical distribution planning. Minimizing the total cost of the objective function with technical constraints make it a combinatorial problem which should be solved by powerful optimization algorithms. In this paper, a new improved hybrid Tabu search/particle swarm optimization algorithm is proposed to optimize the electric expansion planning. The proposed method is analyzed both mathematically and experimentally and it is applied to three different electric distribution networks as case studies. Numerical results and comparisons are presented and show the efficiency of the proposed algorithm. As a result, the proposed algorithm is more powerful than the other algorithms, especially in larger dimension networks.
Ecology and the Environment, May 4, 2007
Methods of modeling the hydrologic process range from human observers to sophisticated surveys an... more Methods of modeling the hydrologic process range from human observers to sophisticated surveys and statistical analysis of climatic data. In the last few years, researchers have applied computer programs called Neural Networks or Artificial Neural Networks to a variety of uses ranging from medical to financial. The purpose of the study was to demonstrate that Neural Networks can be successfully applied to hydrologic modeling. The river system chosen for the research was the Big Thompson River, located in North-central Colorado, United States of America. The Big Thompson River is a snow melt controlled river that runs through a steep, narrow canyon. In 1976, the canyon was the site of a devastating flood that killed 145 people and resulted in millions of dollars of damage. Using publicly available climatic and stream flow data and a Ward Systems Neural Network, the study resulted in prediction accuracy of greater than 97% in +/-100 cubic feet per minute range. The average error of the predictions was less than 16 cubic feet per minute. To further validate the model's predictive capability, a multiple regression analysis was done by Dr. A. Kader Mazouz on the same data. The Neural Network's predictions exceeded those of the multiple regression analysis by significant margins in all measurement criteria.
One of the objectives in conducting this work is to develop a process optimization for manufactur... more One of the objectives in conducting this work is to develop a process optimization for manufacturing. Screen printing is the first and one of the most critical process steps in a printed circuit board assembling manufacturing. Solder paste volume on each pad has a direct effect on the product quality and reliability. This experiment concentrates on solder paste volumes obtained on fine pitch pad geometry's. Pads with this geometry have a 0.020" pitch between lead centers. The key is to have just the right volume deposited in the pad. Too little will cause a faulty joint or an unreliable joint with an electrical continuity failure potential. On the other hand, too much will create solder short problems or worse yet, solder balls. Solder balls can become detached after the product has been tested, causing potential field failures. A 2 4 factorial design has been considered in order to optimize the process.
AAU Journal of Business and Law, 2017
Innovation and leadership are so closely interrelated that leadership thinks of the future and in... more Innovation and leadership are so closely interrelated that leadership thinks of the future and innovation is the fruit of leadership in the future. This paper aims to determine whether leadership leads to innovation. Under visionary leadership, innovation can progress and bear fruit. An appropriate framework to provide a platform for innovation is a necessary condition and requires the appropriate leadership to guide its flow in that direction. Competency for innovation arrises from two main cores: interior and exterior. The exterior core has three components: culture, resources, and habitat; the interior core also has three components: knowledge, attitude, and imagination. The components of, the exterior core, relates to the outside world and one's surroundings whereas the components of the interior core relate to us as individuals. The job of leadership is to create a platform wherein the inner forces synchonize with the outside forces. This paper uses Seelig's Innovation Engine, model to measure innovation progress in multiple industries in the Gulf region and to identify the most innovative industries. A survey, distributed to leaders in several major industries, was collected, analyzed, and presented.
Establishing a link between Course Learning Outcomes (LOs) and Program Outcomes (POs) while asses... more Establishing a link between Course Learning Outcomes (LOs) and Program Outcomes (POs) while assessing the course contents and delivery are among the most challenging issues in Higher Education. In the present study two forms were generated based on specific Course Learning Outcomes identified in the syllabus at the beginning of the teaching term: a Student outcome evaluation form and a Faculty outcome evaluation form. The objective is was to assess if the outcomes specified in the syllabus are being delivered and are being delivered throughout the term. At the end of the semester, a student survey was given to students to evaluate the course outcomes. In addition, the faculty evaluated the course outcomes. A matrix was developed mapping the results of the student, the faculty and each assessment contributing to the specified outcome, all are on a similar scale. A mapped matrix was then generated based on the results. The results from the mapped Matrix pinpoint which assignment contr...
The purpose of this study is to determine whether cash flow impacts business failure prediction u... more The purpose of this study is to determine whether cash flow impacts business failure prediction using a neural network. Using a matched sample of 114 failed and 114 non-failed manufacturing firms selected from the Compustat database, accrual-based and cash flow-based neural network models were developed utilizing financial ratios as input variables. A Z-test was performed to test for significance of any difference at the 0.05 level between the classification results of the two models. The accrual-based model correctly classified 92.55% of firms overall in a training sample and 77.5% of firms overall in a holdout sample. The cash flow-based model correctly classified 94.15% of firms overall in a training sample and 82.5% of firms overall in a holdout sample. While the cash flow-based neural network model outperformed the accrual-based neural network model, results of the Z-test revealed that the difference in classification accuracies between the two models was not significant. This ...
SPIE Proceedings, 1992
ABSTRACT
The purpose of this study is to determine whether cash flow impacts business failure prediction u... more The purpose of this study is to determine whether cash flow impacts business failure prediction using a neural network. Using a matched sample of 114 failed and 114 non-failed manufacturing firms selected from the Compustat database, accrual-based and cash flow-based neural network models were developed utilizing financial ratios as input variables. A Z-test was performed to test for significance of any difference at the 0.05 level between the classification results of the two models. The accrual-based model correctly classified 92.55% of firms overall in a training sample and 77.5% of firms overall in a holdout sample. The cash flow-based model correctly classified 94.15% of firms overall in a training sample and 82.5% of firms overall in a holdout sample. While the cash flow-based neural network model outperformed the accrual-based neural network model, results of the Z-test revealed that the difference in classification accuracies between the two models was not significant. This study does not provide evidence that cash flow improves business failure prediction.
One of the objectives in conducting this work is to develop a process optimization for manufactur... more One of the objectives in conducting this work is to develop a process optimization for manufacturing. Screen printing is the first and one of the most critical process steps in a printed circuit board assembling manufacturing. Solder paste volume on each pad has a direct effect on the product quality and reliability. This experiment concentrates on solder paste volumes obtained on fine pitch pad geometry’s. Pads with this geometry have a 0.020” pitch between lead centers. The key is to have just the right volume deposited in the pad. Too little will cause a faulty joint or an unreliable joint with an electrical continuity failure potential. On the other hand, too much will create solder short problems or worse yet, solder balls. Solder balls can become detached after the product has been tested, causing potential field failures. A 2 4 factorial design has been considered in order to optimize the process.
2002 IEEE International Conference on Industrial Technology, 2002. IEEE ICIT '02., 2002
Many research papers have been tried to optimize the parameter and tolerance designs. Parameter a... more Many research papers have been tried to optimize the parameter and tolerance designs. Parameter and tolerance design are the main key to achieve high quality of products. However, high cost still remain in the process due to reducing the tolerance limit. The objective of this work is combines the optimization of parameter and tolerance design in one stage in to the cost function. This cost function is the sum of tolerance cost function and quality loss function. Taylor expansion will be applied in the genetic algorithm, is considering close attention regarding its potential as a novel optimization technique for searching the optimization value.
Computer-aided Engineering Journal, 1991
Basically, an assembly process is a sequence of positioning and fixing in position between the pa... more Basically, an assembly process is a sequence of positioning and fixing in position between the parts of the system to be assembled. However, finding this sequence automatically remains a very complex procedure because of the combinational aspect of the problem. The conventional programming methods of generating the ordered sequence cannot satisfy the modern factory requirements. The new approach, which seems more promising, considers artificial intelligence techniques and particularly the knowledge-based systems concept. This article presents the design of an expert system for assembly planning, with a description of all its components.
... Approved: n Chairperson \Date Bassam AbuAl-Foul, Ph.D. Committee Member Pddro Pellet,Ph.D. Co... more ... Approved: n Chairperson \Date Bassam AbuAl-Foul, Ph.D. Committee Member Pddro Pellet,Ph.D. Committee member Date D*ate' / Russel^Abratt-r—^hTlK Associate Bean of Internal Affairs Date i^./^reston yjines, DBA Date cecutive Associate Dean, H. Wayne Huizenga School ...
... OF THE MULTINATIONAL CORPORATIONS IN ECONOMIC DEVELOPMENT FOR COUNTRIES WITH LIMITED RESOURCE... more ... OF THE MULTINATIONAL CORPORATIONS IN ECONOMIC DEVELOPMENT FOR COUNTRIES WITH LIMITED RESOURCES By Nabeel N. Sawalha There is ... Tammie and Kids, My wife Tammie, sons; Omar and Laith were very patient and emotionally supportive, paying little ...
Uploads
Papers by abdelkader mazouz