Search

Software Defined Networking and other more sophisticated network infrastructure management have improved network programmability, flexibility, and cost-effectiveness over traditional approaches. By separating the network into layers, Software Defined Networking (SDN) technology simplifies network management. SDN networks are physically dispersed but logically centralized. Due to its fragmented structure, SDN is susceptible to different types of attacks. Distributed denial of service (DDoS) attacks is particularly problematic. The Controller is the main target of the attackers; once the Controller is affected, the entire network will shut down. Due to the prevalence of fake IP addresses, detecting DDoS attack traffic on SDN is more challenging. Therefore, effective and precise DDoS attack detection on the SDN is essential. This paper proposes a DDoS attack detection model on SDN leveraging the LightGBM-DDoS algorithm. The LightGBM-DDoS algorithm was benchmarked with XGBoost, Logistic Regression and K-Nearest Neighbours. The experimental results demonstrate that LightGBM-DDoS is superior to other approaches, with an accuracy rate, detection rate, F1-score, and error rate of 0.9972, 0.9940, 0.9970, and 0.0060, respectively. [ABSTRACT FROM AUTHOR]

This paper is conducted to develop a system that can aid universities in tracking students' academic progress by accurately determining dropout rates using machine learning techniques. This has been done by collecting and organizing data into machine readable forms, using predictive methods that determines significant patterns and trends, constructing different models involving machine learning techniques: Random Forest, Logistic Regression, and Neural Network, for forecasting and predictions, and finally interpreting the statistical results of the model. After the evaluation, it has been concluded that the Random Forest achieved the highest accuracy of 95.00%, precision of 97.78%, and F1 score of 96.70%. Despite these results, the researchers recommend developing a more interesting and engaging prototype GUI and utilizing a more balanced and diverse dataset that will enable higher accuracy and deeper understanding of the results. [ABSTRACT FROM AUTHOR]

It is important that multidisciplinary community of scientists first investigate all the myriad potential directions that quantum machine learning might go to discover what it has to offer. Among these avenues, specified algorithms such as Variational Quantum Algorithms (VQAs) have been applied in different industries for simulation, optimization, and prediction purposes. The focus of this paper is to determine the principles associated with the use of VQAs in chemistry, machine leaning, and optimization, amongst many other developed algorithms. The key results indicated that VQAs can successfully be applied in the above-mentioned fields, with the goal to (1) find ground and excited energy states of different molecules in chemistry; (2) maximizing or minimizing an objective function under specified constraints in optimization; (3) developing trainable quantum models for accurate predictions of unknown and unseen datasets. After statistical analysis, it was discovered that the proportion of publications of VQAs for optimization, machine learning, and chemical purposes have been considerably high over the past 6 years. This is justified by the sparking attention that parametrized circuits have offered over the recent years. After comparing different size effects, it was discovered that the fields of application of VQAs are interconnected, since they presented a small effect size, with Cohen’s d value being less than 0.2. From these observations, it can be concluded that the lessons developed from one application can serve as direction for others. The metallurgical industry as a field of chemistry can benefit from the methods developed in VQAs for optimization, simulation, and development of novel materials. [ABSTRACT FROM AUTHOR]

Order picking is the process of collecting products from the warehouse to fulfill specific customer orders, which is known to be labor- and time-intensive. Order picking significantly determines the warehouse's performance, contributing 50% to 75% of operational costs. Therefore, the efficiency of the order-picking process also affects the supply chain performance. This study investigates two of five main order-picking problems jointly, which are classified in its day-to-day operational problem: order batching and picker routing. The algorithm to solve the problem is developed using metaheuristics, the simulated annealing algorithm. The algorithm avoids local optimization to find better solutions. The algorithm is validated through preliminary testing against problem instances of online grocery shopping cases with 1560 products where each order can consist of many items. The proposed algorithm has proven to be more efficient than the previous approach, which gives quality solutions using less run-time, especially for larger problem instances. [ABSTRACT FROM AUTHOR]

This paper is conducted to develop a system that can aid universities in tracking students' academic progress by accurately determining dropout rates using machine learning techniques. This has been done by collecting and organizing data into machine readable forms, using predictive methods that determines significant patterns and trends, constructing different models involving machine learning techniques: Random Forest, Logistic Regression, and Neural Network, for forecasting and predictions, and finally interpreting the statistical results of the model. After the evaluation, it has been concluded that the Random Forest achieved the highest accuracy of 95.00%, precision of 97.78%, and F1 score of 96.70%. Despite these results, the researchers recommend developing a more interesting and engaging prototype GUI and utilizing a more balanced and diverse dataset that will enable higher accuracy and deeper understanding of the results. [ABSTRACT FROM AUTHOR]

Training models in machine learning often requires big computational capacities, even more when it involves Big Data. The Patient Rule Induction Method (PRIM), introduced as a bump hunting algorithm and a subgroup discovery method, already consumes a lot of computing resources when searching for the rules. In this computational research area, several methods have been developed to handle the problem of computing a big dataset: parallel learning, distributed learning and federated learning. In this paper, we investigate the literature review of these three methods to define each one and the differences between them, and then we determine which of these methods is best suited for the implementation of PRIM in big dataset for out futur works in Big Data. [ABSTRACT FROM AUTHOR]

The impact of e-commerce and the ever-increasing integration of the online world to the daily lives of citizens are the driving reasons for the creation of this thesis. To help streamline the user's shopping experience, the E-Commerce Cheapest Clusterer (ECCC) achieves this goal by doing its namesake: to group similar products into corresponding clusters and select the most affordable product in each of the clusters. The result is then displayed in an easy-to-browse interface that directly connects to the project's target website, Lazada. The images are compared using the Structural Similarity Index Measure (SSIM). If the value returned by the SSIM function is greater than a certain threshold, the image enters the same cluster as the one it was compared to. To ensure efficiency, this paper also investigates the results of different threshold values. Setting a lower threshold value resulted in the algorithm finishing faster at the cost of several misplaced images in clusters, whereas a higher threshold value gave the program a considerable runtime but was less prone to misidentification. Additionally, the testing also revealed that setting the threshold value too high can hinder the program's ability to cluster similar images even if the difference is slight. [ABSTRACT FROM AUTHOR]

It is important that multidisciplinary community of scientists first investigate all the myriad potential directions that quantum machine learning might go to discover what it has to offer. Among these avenues, specified algorithms such as Variational Quantum Algorithms (VQAs) have been applied in different industries for simulation, optimization, and prediction purposes. The focus of this paper is to determine the principles associated with the use of VQAs in chemistry, machine leaning, and optimization, amongst many other developed algorithms. The key results indicated that VQAs can successfully be applied in the above-mentioned fields, with the goal to (1) find ground and excited energy states of different molecules in chemistry; (2) maximizing or minimizing an objective function under specified constraints in optimization; (3) developing trainable quantum models for accurate predictions of unknown and unseen datasets. After statistical analysis, it was discovered that the proportion of publications of VQAs for optimization, machine learning, and chemical purposes have been considerably high over the past 6 years. This is justified by the sparking attention that parametrized circuits have offered over the recent years. After comparing different size effects, it was discovered that the fields of application of VQAs are interconnected, since they presented a small effect size, with Cohen's d value being less than 0.2. From these observations, it can be concluded that the lessons developed from one application can serve as direction for others. The metallurgical industry as a field of chemistry can benefit from the methods developed in VQAs for optimization, simulation, and development of novel materials. [ABSTRACT FROM AUTHOR]

Most of the items managed by a retailer are perishable and have a limited shelf life. The longer they are stored, the more their value decrease, where the holding cost is directly proportional to the value of the goods and deterioration is time proportional. If there is a shortage, it will be resolved backorder. The supplier allows payments within a certain period of time. This paper will try to combine and modify some basic inventory models to develop a model for deterioration items that have expiry date with varying holding cost to time where deterioration is time proportional, under the conditions of backlogging and allowing delays in payments. The model solution is carried out with an optimization approach based on the parameters that affect the model. The approximate optimal solution has been obtained. Numerical examples are given at the end of this paper to illustrate the model settlement algorithms. [ABSTRACT FROM AUTHOR]

This paper buttresses the need for IoT-enabled agriculture, in order to tackle food insecurity globally, especially in the Sub-Saharan African (SSA) region where the global population rise is predicted to concentrate. It details the nittygritty involved in setting up a smart irrigation system and presents a systematic review of seven (7) scheduling algorithms used to drive smart irrigation systems. Analysis of the algorithms shows that they differ in terms of sensor inputs and level of precision. Recommendations were made in choosing a suitable algorithm and how the chosen algorithm can be deployed to withstand network connectivity problems. This paper also proposes a community Internet of Things (IoT) network to deploy smart irrigation systems and other applications of IoT-enabled agriculture, in rural areas among smallholder farmers. This is to mitigate the challenges that are likely to discourage farmers especially those in the SSA from embracing the technology. [ABSTRACT FROM AUTHOR]

The Vehicle Routing Problem (VRP) is a problem related to the distribution of products using the optimal and involving more than one vehicle by taking into account several obstacles in serving a number of agents according to the request of each related agent. The Capacitated Vehicle Routing Problem (CVRP) is one of the variations of VRP, namely by adding constraint in the form of vehicle capacity used in product distribution. This paper applies the formation of a CVRP model to the problem of the distribution route of the Solopos daily mail and its solution using the Sweep Algorithm method which aims to optimize the distribution route. The Sweep Algorithm is an algorithm consisting of two stages, the first stage is clustering agents, then the second stage is the formation of routes for each cluster with the Nearest Neighbor method. Based on the calculations carried out in solving CVRP problems using the Sweep algorithm, the results obtained that the total vehicle mileage is 216,5 km with a time windows of 6 hours 5 minutes. [ABSTRACT FROM AUTHOR]

In most of the imaging applications the camera is placed in outdoor conditions with the lens open to the environment. Example, cameras installed on the buildings for surveillance, cameras fitted on cars that are open to the environment for ADAS applications, camera for flight passengers to view outside scenes on the fly etc. One of the most challenging problems in such scenarios is that the dust particles may easily form a partially or fully occluding layer on the camera lenses and block vision. There are more challenging scenarios like the presence of wind, rain, mist, sun glare etc. which can also lead to false imaging algorithm results. The purpose of this paper is to solve this problem efficiently by detecting dirt/dust if presents on the lens. This is helpful to stop the system functioning instead of providing misleading results. This application is more essential for safety critical applications like ADAS where the accurate real time detection of objects must happen. Proposed approach has more than one algorithm like edge analysis, inter frame analysis, edge strength analysis etc. to produce accurate result for dust/dirt detection. [ABSTRACT FROM AUTHOR]

In most of the imaging applications the camera is placed in outdoor conditions with the lens open to the environment. Example, cameras installed on the buildings for surveillance, cameras fitted on cars that are open to the environment for ADAS applications, camera for flight passengers to view outside scenes on the fly etc. One of the most challenging problems in such scenarios is that the dust particles may easily form a partially or fully occluding layer on the camera lenses and block vision. There are more challenging scenarios like the presence of wind, rain, mist, sun glare etc. which can also lead to false imaging algorithm results. The purpose of this paper is to solve this problem efficiently by detecting dirt/dust if presents on the lens. This is helpful to stop the system functioning instead of providing misleading results. This application is more essential for safety critical applications like ADAS where the accurate real time detection of objects must happen. Proposed approach has more than one algorithm like edge analysis, inter frame analysis, edge strength analysis etc. to produce accurate result for dust/dirt detection. [ABSTRACT FROM AUTHOR]

Websites have different purposes. Some of which intend legitimate functions in the economy while some of which intend harmful cases towards users. Although various research has been made to address this problem, these detection systems still leave plenty of room for improvement, specifically on its performances. This study was based on the recommended approach for future work by a related work wherein it contains 10 base features of a URL for its classification. The recommended approach states that an extended number of features from the base features increases the detection accuracy. In this paper, it proposes a comparison between the performance of three cases: the base 10 features, an extended feature set, and a set where a feature selection algorithm is applied. The researchers utilized machine learning algorithms to build models in classifying legitimate and malicious URLs. The study showed that there is a directly proportional relationship with a model's number of features and a model's performance. Extending the number of features of the data set leads to an increase with the performance of each model. [ABSTRACT FROM AUTHOR]

Optimizers are approaches or algorithms dependent to enhance the characteristics of the Neural Network (NN) like weights and learning rate in order to decrease the loss rate, On the other hand, Siamese Neural Network (SNN) are two identical sub-networks, they work in parallel and they are sharing parameters and weight, SNN uses for indicate similarity. In this research, we study the effect of optimizers Siamese Neural Network, using Digits handwritten (MINST) dataset, the effects is studied for Adam, Nadam, Adadelta and SGD optimizers with respect to process time and accuracy, the accuracy is 97%, 97%, 79% and 92%. [ABSTRACT FROM AUTHOR]

Real-life optimization problems consist of various conflicting optimization problems. Due to the interdependency of these sub-optimization problems, their solution is more difficult than single optimization problems. For this reason, most of the real-life optimization problems are considered multi-objective optimization problems by nature, and different methods and approaches are applied. In this paper, a recent multi-objective optimization problem, the bi-objective travelling thief problem (BiTTP) was studied. Two state-of-the-art algorithms, ALNS and NSGA-II, were hybridized with certain local search methods and applied on six problem instances of Polyakovskiy's problem benchmark set. As a result, the performances of these two algorithms were evaluated and a new set of Pareto front sets were generated for the related benchmark instances. [ABSTRACT FROM AUTHOR]

Metaheuristic methods have many design parameters, and fine-tuning of that parameters can improve these algorithms performance. In this paper, a sampling based algorithm configuration approach is proposed and applied it to the Differential Ant-Stigmergy Algorithm (DASA)'s five control parameters. Performance of a large parameter set of DASA, obtained by Latin Hypercube Hammersley Sampling (LHHS) method, used to solve the Sphere function and compared it with another tuned version. The results of our experiment demonstrated that LHHS method found better performing configurations than the default parameter value of the DASA and also than another proposed tuned version DASA*. And the results demonstrated that three parameter configurations obtained with LHHS found better result than the best configuration obtained with Sobol Sequence Sampling method (DASA*) for function dimension 20, and five parameter configuration for function dimension 40. According to the results, it can be said that usage of LHHS for initialization of other state-of-art algorithm configuration methods instead of other sampling methods is worth investigating. [ABSTRACT FROM AUTHOR]

Industrial Structures such as pipes develop issues over time which are primarily due to ageing, corrosion, cracks, or other forms of mechanical defects which could potentially lead to leakages or further uncontrollable fatal occurrences. As a result, inspections of such industrial structures are therefore of extreme importance. While prevailing inspection techniques are usually sensor based, time-consuming and labour-intensive, the application of robotic systems significantly reduces the human effort required for the same level of inspection. In this paper, implementation of a vision-based approach and a suitable low-cost solution was developed for detecting defects in industrial structures such as pipelines and hollow cylindrical structures. The camera embedded in the robot is used to identify defects such as cracks, corrosion and blockage in real time through a systematic algorithm of dataset training (1500 datasets) representing these conditions. This paper also focuses on the modelling and analysis of the dataset training of a pipeline inspection robot. The machine vision system thus developed was capable of effectively detecting defects even in low light conditions with a mean average precision of 91%. [ABSTRACT FROM AUTHOR]

One of the problems faced in production facilities is the material handling, which may account between 20-50% of total costs of manufacturing and distribution. Material handling is directly related to facility layout. In many instances, wrong facility layouts result in excessive handling costs, which could be eliminated or minimized through effective layout procedures. In this paper, we present a case application of a facility layout for a furniture factory, which has a wood and a fabric sections, with fourteen and fifteen departments respectively. About 23 types of sofa models are produced in the factory, where excessive material and personnel flows and backflows were observed during a detailed study. Facility layout models were used to estimate total material flow for both the current and the improved layouts. It was found that by using appropriate layout techniques, unnecessary flows could be eliminated, and total material flow costs could be minimized. Interdepartmental distance matrices, material and worker flow matrices, and weighting factors were used to quantify the total costs of flow between departments. By using quantitative relations and a facility layout algorithm in excel, layouts of departments were restructured, and total flows were minimized. The proposed new layouts resulted in a significant reduction in material handling costs. [ABSTRACT FROM AUTHOR]

One of the problems faced in production facilities is the material handling, which may account between 20-50% of total costs of manufacturing and distribution. Material handling is directly related to facility layout. In many instances, wrong facility layouts result in excessive handling costs, which could be eliminated or minimized through effective layout procedures. In this paper, we present a case application of a facility layout for a furniture factory, which has a wood and a fabric sections, with fourteen and fifteen departments respectively. About 23 types of sofa models are produced in the factory, where excessive material and personnel flows and backflows were observed during a detailed study. Facility layout models were used to estimate total material flow for both the current and the improved layouts. It was found that by using appropriate layout techniques, unnecessary flows could be eliminated, and total material flow costs could be minimized. Interdepartmental distance matrices, material and worker flow matrices, and weighting factors were used to quantify the total costs of flow between departments. By using quantitative relations and a facility layout algorithm in excel, layouts of departments were restructured, and total flows were minimized. The proposed new layouts resulted in a significant reduction in material handling costs. [ABSTRACT FROM AUTHOR]

Recent supply chain targets have led to decreased response time requests also causing enterprises invest for automation technologies for efficient management for their warehouses. In this paper, we focus on an automated warehousing environment where there are robotic order pickers travelling on ground flexibly throughout aisles. We develop deadlock and collision prevention algorithms for safely travel of those robotic order pickers. The developed algorithms can be utilized by any warehouse system having autonomous vehicles (e.g., automated guided vehicles, mobile robots, or shuttles in a multi-tier automated warehouse) that can travel between aisles freely. While developing those smart algorithms, we also aim to improve some performance metrics such as average flow time and, maximum flow time of a transaction in the system. To test the performance of the algorithms, we model the system as a multi-agent system, where autonomous vehicles are able to communicate with each other under an Internet of Things (IoT) environment. We compare that flexible system design with a dedicated design where a dedicated vehicle is assigned for a predefined number of aisles. The experimental results show that under well-designed deadlock and collision prevention control algorithm, flexible travel of vehicles can provide advantage compared to dedicated design. [ABSTRACT FROM AUTHOR]

This paper proposes a batch scheduling model where two time-changing effects simultaneously occur, i.e., parts deteriorating and operator learning-forgetting effects. A mathematical model is proposed for the problem where the decision variables are the number of batches, batch sizes, and a schedule of the resulting batches to minimize total actual flow time. A proposed algorithm is developed by trying different numbers of batches, starting from one and then increasing it one by one until the objective function value does not improve anymore. Numerical examples show that the faster operators learn, the lower the optimal number of batches and the lower the optimal total actual flow time. Second, the faster parts deteriorate, the higher the optimal number of batches and the higher the optimal total actual flow time. Finally, the model divides parts into small batches to prevent the processing time from increasing due to part deterioration. [ABSTRACT FROM AUTHOR]

The aim of this study is to determine an optimal new factory location for a company based on the proximity of the new factory's location to its customers and, in turn, reduce the cost related to deliveries. This was carried out using the transportation algorithm via the excel solver tool. This takes into account variables such as distance from customers, weight of the goods, as well as vehicle-based and driver-based costs, in order to determine the most cost effective location of a new factory. The current factory for the company case study is based in Pinetown and Kwa-Zulu Natal in South Africa. The proposed factory locations are Pietermaritzburg, Durban Central or another in Pinetown, all within Kwa-Zulu Natal. Based on the results, the optimal location to establish a new factory would be Pietermaritzburg, KwaZulu Natal. This location combined with the current factory location will generate the lowest cost with regards to vehicle-based and driver-based costs when delivering goods. [ABSTRACT FROM AUTHOR]

The aim of this paper is to provide ANOVA study for thermal profile process of manufacturing Printed Circuit Boards (PCB). The thermal profile process controls the cycle time and temperature level of the oven, in other words, the solder joints have to reach the correct cycle time of the soldering temperature. Therefore, it is challenging to simultaneously control the temperatures in different zones with specific cycle time in each zone. The solder paste and components should be well-matched in the heating slope for each zone; the solder paste has different types and different manufacturers. Also, these variables (temperature and time in each zone) are interconnected and thus, result in good quality in solder joint; they also determine which variables are more significant to get good solder quality by using design of experiment (DOE). Minitab software will be used for DOE. The data of the research are collected from real assembly line of PCB. Obtaining a better yield via improving the time of the PCB assembly process is another aim of the research. Hence, the outcome to all of these attempts of optimizing the assembly line of PCB for continuous improvement is to deliver good quality products, reduce cost, and minimize the time of delivery and meet the customer expectations. [ABSTRACT FROM AUTHOR]

Social network theory is becoming more and more significant in social science, and the centrality measure is underlying this burgeoning theory. In perspective of social network, individuals, organizations, companies etc. are like nodes in the network, and centrality is used to measure these nodes' power, activity, communication convenience and so on. Meanwhile, degree centrality, betweenness centrality and closeness centrality are the popular detailed measurements. This paper presents these 3 centrality in-depth, from principle to algorithm, and prospect good in the future use. [ABSTRACT FROM AUTHOR]

Social network theory is becoming more and more significant in social science, and the centrality measure is underlying this burgeoning theory. In perspective of social network, individuals, organizations, companies etc. are like nodes in the network, and centrality is used to measure these nodes' power, activity, communication convenience and so on. Meanwhile, degree centrality, betweenness centrality and closeness centrality are the popular detailed measurements. This paper presents these 3 centrality in-depth, from principle to algorithm, and prospect good in the future use. [ABSTRACT FROM AUTHOR]

Pneumatic conveying of powders is an engineering process used for conveying dry granulate or powder material. The aim of this paper is to present a modeling methodology to optimize energy efficiency of pneumatic conveying systems, considering the specific bulk characteristics of the product being conveyed. This work is based on engineering optimization of a workflow with data from an industrial operation commanded by a Programmable Logic Controller (PLC) with a control algorithm, performing logical, sequential, and timed tasks for plant control. The PLC communicates with a Human-Machine Interface and a Supervision and Control System, which are the means of interaction through a graphical environment interface with the process operator. By applying mathematics to introduce a systematic method to select the gas (air) pressure and flow necessary to operate a pneumatic conveying system in dense phase, it has been shown, on an industrial scale, the feasibility of controlling a conveying system by manipulating only two input parameters. This allows operation at pre-determined conveying rates with lower power requirements, resulting in a reduction of OPEX. Thus, this research focuses on optimizing energy efficiency to reduce Operational Expenditure (OPEX) through a systematic modeling approach. Because pneumatic conveyance is highly empirical, general models are difficult to establish. Due to these limitations, evaluating energy efficiency is usually limited to a specific experimental range of conditions. This methodology can be adapted to enhance the energy efficiency of other types of pneumatic conveying systems. [ABSTRACT FROM AUTHOR]

Vehicular Networks are likely to be implemented in future vehicles. Their usage can enhance road safety and enable new forms of communications such as Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communications. The simulation of this type of network is the most useful method to test new techniques, algorithms, and architectures. This is basically due to the high cost of deploying such systems in real scenarios. Named Data Networking (NDN) is a new architecture that researchers have proposed recently as a promising architecture for future Internet technologies. It is expected to support various applications in vehicular communications and aims to overcome the problem of classically TCP/IP-based architecture concerning mobility, by relying on naming the content rather than using end-to-end node addresses. Indeed, the in-network caching feature in NDN makes this architecture more suitable for vehicular networks. To prove the usefulness of NDN in connected vehicles, we propose a simulation study in an urban network architecture. [ABSTRACT FROM AUTHOR]

The industrial sector globally is at the cusp of fourth industrial revolution. "Industry 4.0" first introduced in 2013, focuses on the concept of smart factory and cyber physical systems thus integrating advanced technologies such as automation, cloud computing, internet of things, 3-D printing, blockchain and big data analytics. As its extension in one of the prominent service sectors, the soul of "Healthcare 4.0" lies in the integration and convergence of real and virtual worlds, wherein manual appointments, in-person examinations and treatment systems can be integrated with software, algorithms and people. Emerging technology applications such as wearables coupled with the artificial intelligence and predictive diagnostics capabilities have already started getting traction, globally. However, the policy interventions by governments to adapt the structure of medical policies in view of evolving technologies and data security concerns are still lacking. The need of the policy alignment becomes imperative especially in case of India, which is emerging as a medical tourism hub. This paper presents a concept note while capturing the emerging trends in Healthcare 4.0 from an Indian perspective. Also, the mapping of possible health applications till date has been done to analyze the evolution of the area across multiple disciplines. Current research issues and challenges for the implementation are further highlighted. The research trend in this domain is expected to increase at least for the next decade. [ABSTRACT FROM AUTHOR]

This paper extends inventory models for growing items by considering quality inspections, permissible shortages with complete backordering, and holding cost during both the growth period and the consumption period. By backordering shortages, the firm can avoid the loss of sales by paying delay penalties to customers who wait for late items. After items reach certain weight preferred by the customers, they are fully inspected, and inferior-quality items are removed at the end of inspection period. The model determines the optimum cycle length and shortage level to minimize the total cost of the inventory system. This cost includes the purchasing, setup, inspection, feeding, holding, and shortage costs. A nonlinear programming model is formulated and an optimum solution algorithm is presented. [ABSTRACT FROM AUTHOR]

The technical solution used to improve and manage the storage and retrieval of data from databases is known as a database management system (DBMS). A transaction signifies a change in the database's data. It is an activity or sequence of operations performed by a single user or application software to read or change the database's contents. A transaction can fail before all the actions in the set are completed, which is a concern. This might occur because of a power outage, a system crash, or other factors while Concurrency Control in a Database Management System is a method of controlling many actions at the same time without their interfering. To resolve these issues, the various techniques are presented in this study with discussion of techniques and algorithms used. Concurrency control methods like locking, timestamps, and Optimistic Concurrency Control can be used to handle concurrent transactions and avoid conflicts. Studying a distributed database for relational (MySQL) and NoSQL (MongoDB), as well as the implementation and performance of ACID transactions in NoSQL, is another proposal (MongoDB). [ABSTRACT FROM AUTHOR]

This paper investigates the use of elitism in chemical reaction optimization (CRO) to address its convergence performance in difficult problems. We focus on problems with complex and highly discontinuous solution space. In such problems CRO’s convergence performance tend to be sluggish as the algorithm repeatedly digress from the bestfound solution characteristics in search for solutions in different areas in the problem’s solution space. A complex road network design problem is used to demonstrate this issue and experiment with the impact of elitism on algorithm convergence. Elitism has been used successfully in evolutionary algorithms. Results show that its use in CRO improves algorithm convergence performance drastically. However, due to CRO’s tendency to have diminishing population of molecules in such problems, the use of a larger list of elite solutions appears to be ineffective in improving the algorithm performance beyond the initial gains from introducing elitism. We investigate the reasons behind this observation and point to possible solutions. [ABSTRACT FROM AUTHOR]

This paper investigates the use of elitism in chemical reaction optimization (CRO) to address its convergence performance in difficult problems. We focus on problems with complex and highly discontinuous solution space. In such problems CRO’s convergence performance tend to be sluggish as the algorithm repeatedly digress from the bestfound solution characteristics in search for solutions in different areas in the problem’s solution space. A complex road network design problem is used to demonstrate this issue and experiment with the impact of elitism on algorithm convergence. Elitism has been used successfully in evolutionary algorithms. Results show that its use in CRO improves algorithm convergence performance drastically. However, due to CRO’s tendency to have diminishing population of molecules in such problems, the use of a larger list of elite solutions appears to be ineffective in improving the algorithm performance beyond the initial gains from introducing elitism. We investigate the reasons behind this observation and point to possible solutions. [ABSTRACT FROM AUTHOR]

This paper presents the performance comparison of five meta-heuristic algorithms to solve a discrete optimization problem. The comparison is undertaken for a case of simply supported plate subjected to biaxial loading conditions. Furthermore, the optimization objective is to determine the optimal stacking sequence design of a laminate that maximizes the critical buckling load factor (cb). The chosen metaheuristics have been implemented using MATLAB with the same convergence criteria and the same maximum number of iterations to ensure a fair comparison. The implemented assessment criterion has performance measures of average CPU time, solution price, reliability, and normalized price. The results have demonstrated the outperformance of the Ant Colony Optimization Algorithm (ACOA) over other algorithms, which confirms the findings of previous studies. Moreover, the Tabu search algorithm (TS) and the Discrete Particle Swarm Optimization algorithm (DPSO) performed poorly due to their limited exploration capability. Additionally, the Genetic Algorithm (GA) and the Simulated Annealing algorithm (SA) exhibited a high level of reliability but showed an expensive solution cost. This study presents an adequate comparison approach of meta-heuristics, where it extends the comparison scope to cover the performance analysis of meta-heuristics more than that previously done in the domain of stacking sequence design optimization. [ABSTRACT FROM AUTHOR]

The objective of this paper is to determine the current methods and data sources used to model mode choice for transportation of the population in universities and urban areas. We performed a systematic literature review to compare the different methods, techniques, data sources, and metrics used to model mode choice behavior and evaluate its performance. A mode choice model is a discrete choice model, which predicts a choice in a set of discrete alternatives; it describes the selection of a transportation mode from an individual, in disaggregated models, or a percentage of a population, in aggregated models. We collected a total of 60 primary studies after searching on different databases from 1982 to 2021, most of them being published in journals. The studied areas range from small regions, such as universities, to wider regions, like metropolitan areas. As most of the models in the studied research papers used data collected from surveys the authors created for that specific research, we advise to ensure that the data sources for the model training contain the required sample characteristics and variables. If such database is not available, performing a survey as an initial data gathering phase is recommended. Another finding is that mode choice models can be categorized as econometric, in which parameter estimation is interpretable, and black box models, in which parameters are not explicit, therefore practical interpretations about the relation of each variable with the mode choice probability cannot be made. We conclude that applying econometric models is still recommended for initial experiences developing discrete choice models, as they continue to be widely used. Given that they are composed of a utility function, followed by the application of a model, such as the Logit and its variants, a recently used option consists of incorporating black box algorithms for the utility calculation, enhancing the quality of predictions, and then applying econometric models to keep parameter interpretability. The study and application of mode choice models have increased, having a growing number of publications in recent years. Additionally, applied algorithms have evolved and range from specialized artificial neural networks such as the deep neural networks, to classification algorithms, such as the random forest classifier. Therefore, we expect the relevance of this research area to continue growing as technology and new methods develop. [ABSTRACT FROM AUTHOR]

Industrialization is growing rapidly in line with increasing needs from consumers, making competitions stronger. To continue developing and competing with competitors, a company must increase the productivity in order to meet consumer needs. PT. XYZ produces metal fitting products with make-to-order principle. This study aims to plan and control the number of raw materials used in production to minimize costs without reducing the quality of the products. The Material Requirement Planning (MRP) method is recommended because it can help raw material procurement to be more effective and efficient. Moreover, implementing safety stock system anticipates sudden fluctuations in demand. Thus, the company will not lose the opportunity to meet consumer needs as scheduled in the first place with minimum possible cost. In this case, the use of MRP method with Wagner Within Algorithm (AWW) lot-sizing technique will result in the lowest costs compared to Economic Order Quantity (EOQ), Period Order Quantity (POQ), Lot for Lot (LFL), Silver Meal Algorithm, and Least Unit Cost (LUC). Using the AWW lot-sizing, the order and storage costs of raw materials will be saving Rp1,285,135 with a percentage of 19% compared to the usual method PT. XYZ usually used. [ABSTRACT FROM AUTHOR]

PT Peace Industrial Packaging is a company that manufactures various types of packaging in Styrofoam, Blow Molding Bottles, PET Bottles, and Injection Cap. The company uses a make-to-order system to fulfil every customer request. Therefore, a good layout is important in supporting the entire production process to run effectively. However, the company is currently implementing a process layout that has not maximized production flow, requiring relatively large travel time, transportation distance, frequency, and cost. This research was conducted to design the improved layout as the best alternative layout solution with optimal flow movement using the SLP (Systematic Layout Planning) method, the CRAFT algorithm (Computerized Relative Allocation of Facilities Techniques), and CORELAP (Computerized Relationship Layout Planning). The final results of the research are calculating the flow of material movement to finished products. These material handling costs consider the distance travelled, frequency and time of movement and compare the layout results of the three methods to obtain a more effective layout design. The most effective layout is the SLP method in the second design which provides a distance reduction of 60.21% and material handling costs 44.29%. The layout implementation will be visualized using Promodel and Flexsim simulation software. [ABSTRACT FROM AUTHOR]

In this paper the multi objective optimization problem is studied further with respect to quality indices. In particular, the maximum Pareto front error, accuracy of Pareto frontier, spacing, overall spread, objective spread, maximum spread, and number of distinct choices, cluster, hyper area, dominant area, nondominated evaluation, and crowding distances are employed to assess the Pareto Front of real case studies. Two algorithms are employed to solve the multi objective problem, namely: the NSGA II and SPEA2 algorithms. These performance indices are analyzed in relation to a set of benchmark problems and conclusions are drawn. Five out of twelve indices can be eliminated. In addition, a new approach to handle both constrained and un-constrained multi objective problems using a modification to the fast elitist multi objective Genetic Algorithm, NSGA II (Non-dominated Sorting Genetic Algorithm II). The approach overcomes shortcomings from previous approaches in handling constrained multi objective problems like Penalty Function Approach, Jimenez-Verdegay-Gomez-Skarmeta's approach, and Ray-Tai-Seow's Method. Comparative studies are conducted using a group of quality indices [2,19] to compare the new approach (M-NSGA II) to two other different algorithms: regular NSGA, NSGA-II with penalty function. Results show that the new approach is the best among the three approaches. [ABSTRACT FROM AUTHOR]

Production scheduling is defined as allocating limited resources to do several jobs. PT. Hari Mukti Teknik is a company engaged in manufacturing and engineering in Indonesia. The company implements an FCFS scheduling system with make to order and job shop production system. In order to produce a schedule that is efficient and feasible for the production process in the company, it is necessary to carry out further research on the production scheduling process using the heuristic method. The proposed method is Shifting Bottleneck Heuristic algorithm. This method can determine which machines are experiencing bottlenecks so that the scheduling results in better job sequences and produces a smaller makespan. The current method results in the makespan of 603 hours, while the Shifting Bottleneck Heuristic algorithm has smaller makespan, that is 468.71 hours. The proposed method produces better results than the company's existing method, where EI is 1.287 and RE is -28.651% based on the performance evaluation with EI and RE parameters. Therefore, the Shifting Bottleneck Heuristic algorithm is a better method and can be applied by the company to minimize the makespan. [ABSTRACT FROM AUTHOR]

Kujang (Sundanesse cleaver) as one of the cultural heritages of the Sundanese people (residents of West Java, Indonesia) has various attributes. These attributes can be used as a media for learning culture by the community. Currently the introduction of Kujang is only limited to general information. In fact, each attribute of the Kujang contains a very meaningful cultural philosophy. One of the cultural philosophies on the Kujang is shown in the motif design of the Kujang. This study aims to build a digital promotion media of Kujang through a labyrinth educational game which is implemented using the A* algorithm. The use of the A* algorithm is expected to guide educational game users to find the final destination of the maze by choosing the fastest path. In this educational game, there are five choices of labyrinth paths that formed from five motif Kujang designs. This research shows the novelty that refers to the application of the motif Kujang design as a labyrinth route. The labyrinth route which is formed from five motifs Kujang ie.: tapak nanggala, tirta sadana, nagabandang, waruga sungsang, and hanjuang, which has different route difficulty levels. Based on the results of the usability test, it shows that globally the respondents show a level of comfort in using educational games at the level of 71.72%. The interactive and interesting promotion of the motif Kujang design can be done through this game by providing an experience of exploring the maze paths based on the motif kujang design. The labyrinth education game based on the motif Kujang design is able to perform well in the aspects of control (82.25%), efficiency (70.75%) and effects (70.01%). This game still shows moderate level in learnability performance (63.95%). This research is still limited to the creation of labyrinth paths based on the motif Kujang design and has not been equipped with the strengthening of information on the meaning of other cultural attributes contained in the motif Kujang design. Therefore, the opportunity to develop research towards strengthening information on the meaning of attributes integrated with the appropriate algorithms is very potential. [ABSTRACT FROM AUTHOR]

The sequencing of groups of jobs on a workstation, when each group is periodically released, is a common scheduling practice in manufacturing. When the objective is to minimize the sum of completion time, it becomes an NP-Hard problem. Solution to small instances of this problem, and most NP-Hard problems, can be found using integer programming algorithms such as branch-and-bound or cutting plane strategies. However large instances must settle for an approximate algorithm using some heuristic or dispatch rule. In this paper, we present a composite dynamic dispatch rule named "Gap or No Gap" (GORNG) for tight schedules of large instances of the 1 | rj |ΣCj problem that performs better than other commonly used approaches. Our approach is novel in that it is counter-intuitive to keep a machine idle sometimes when work is available. We prove that in some cases it is better to force idle time rather than to schedule the next available job when the goal is to minimize the sum of the completion times of all jobs. The proposed Composite Dynamic Dispatching Rule (GORNG) was tested using four randomly generated datasets. The results prove that our rule consistently performed better than both the Gap (G) and the No Gap (NG) rule. In the four data sets, GORNG was better 75.00%, 72.73%, 85.16%, and 71.88% of the time with reference to our objective of minimizing the completion time. Since most scheduling scenarios involve future demand that is not always known in advance, our proposed dispatch rule "Gap or No Gap" (GORNG) provides a better strategy with reference to the objective. Our proposed algorithm has important applications in practice. Operations Managers should realize that scheduling all unfinished jobs on hand does not guarantee an optimal solution all the time as there are certain instances where it is better to force idle time rather than scheduling all the unfinished jobs. [ABSTRACT FROM AUTHOR]

Conventionally, decision makers evaluate vendors vis-a-vis traditional criteria (e.g., costs, implementation reliability and quality). However, this approach ignores other important concerns such as resilient implementation and running of a project that is a crucial aspect towards service continuity. This research presents a decision-making approach that helps managers to evaluate vendors profiles based on their submitted tenders to accomplish a proposed radio frequency identification (RFID)-based passport tracking system. First, this evaluation is conducted based on two sets of traditional and resilience criteria presented in a unified framework based on literature and the case company's experts. Second, the weights of identified criteria are quantified by using the decision-making trial and evaluation laboratory (DEMATEL) algorithm. Third, the ELimination Et Choix Traduisant la REalité (ELECTRE) algorithm is applied to evaluate the performance of vendors to recommend the best one. Four employees from the case company participated in evaluating the applicability of the proposed decision-making tool. This includes the selection of one vendor out of 6 submitted their proposals to implement RFID-based passport tracking system. [ABSTRACT FROM AUTHOR]

This study attempted to evaluate evapotranspiration (ET) estimation algorithm at a lodgepole pine tree open forest in eastern Idaho, United States. The coordinate location of study area (Latitude is 44.5o N; Longitude is 111.4o W) and the Elevation is 1950 meter. By performing eddy covariance method for the data collected under 10:30 am in local solar time, clear-sky condition and 16-day period. Furthermore, the accuracy of ET model was evaluated by using sensible flux dataset measured by scintillometer. This experiment considers scintillometermeasured sensible flux data to evaluate the accuracy of ET model. In scintillometer, evapotranspiration computed using radiation measurement data. There were two sets of sensible and latent heat flux data. One was measured on the south tower, and the other was measured on the north tower. The dataset from the south tower was much more reliable in term of data availability. This study used sensible heat and latent heat flux data from the south tower. The dataset from north tower were used only as supplementary dataset when the dataset from south tower were not available. This study concluded that scintillometer was be able to provide ground truth ET dataset to improve the accuracy of ET estimation algorithm and it suggests that to obtain the accuracy measurement of ET estimation actual, a simple linear adjustment should be applied in the ET algorithm model. [ABSTRACT FROM AUTHOR]

The motivation for this work is in its application to manufacturing systems, and we use a manufacturing model focusing on the following three points: efficiently mass-producing the same product; evaluating the number of products per unit time using tact time; and avoiding collision between materials as much as possible. A parallel machines model was presented in a previous paper as such a manufacturing model, and an efficient algorithm to compute collision probability in this model was also proposed. In this work, we focus on the same parallel machines model but with buffer space added. Buffers are spaces for materials awaiting processing and are used in order to avoid a collision within a manufacturing system. We develop an efficient algorithm to compute collision probability in a parallel machines model with buffers. Moreover, in order to show that our algorithm runs fast in reality, we implement the algorithm on a PC and show its performance through computational experimentation. [ABSTRACT FROM AUTHOR]

In this work, the demand pattern of readymade garments of a knit garments factory in Bangladesh was studied and factors influencing the demand pattern were identified. An accurate sales forecast is a prerequisite of achieving an accurate and effective supply plan of complete garments. A Fuzzy Inference System based algorithm was developed to predict the sales of garments according to the sales influencing factors. The developed algorithm is mainly a quantitative method of forecasting but also takes into account some qualitative issues. The developed model was compared with the actual total sales found by traditional seasonal forecast with trend adjustment. The seasonal factors for the two cyclic demand patterns, January -- June, and July - December of each year were calculated and the Linear Regression trend was followed to calculate the traditional forecasting. The developed model showed better prediction as it matches closely with the actual sales. The Root Mean squared Error (RMSE) and Mean absolute deviations (MAD) are 3.605 and 3.081 in the developed fuzzy model which is within the standard limit. Thus the obtained result showed by the Fuzzy model yields better demand prediction taking into account the inherent uncertainties and with less error and thus the efficiency of the proposed Fuzzy model was verified. [ABSTRACT FROM AUTHOR]

Past studies have witnessed the wide application of assisted history matching for the calibration of dynamic reservoir models. Although the proposed algorithms have the potential to improve the history matching process in some synthetic cases, most of them have failed or have partially succeeded when applied to real, complex reservoirs. Thus far, identifying the most efficient optimization strategy for history matching has remained a challenging topic for research. In this paper a sequential approach is adopted whereby a reservoir model is replaced by a proxy model and multi-objective optimization algorithms are applied on misfit functions which was defined by combination of the proxy models and historical data. The proposed approach was tested on a case study involving a benchmark synthetic reservoir model with 14 years of production data. The data was freely provided by Imperial College London. The effectiveness of using individual optimization algorithms were quantified by using normalized root mean square error. The proposed approach is found to be efficient, robust and flexible. [ABSTRACT FROM AUTHOR]

In this paper, we have summarized three inter-related multiple criteria models for power generation expansion planning that we have published recently. These papers also present different solution or algorithmic approaches for solving these models and use the same case study involving realistic data for the power generation expansion planning problem. The models, and their solution approaches, emphasize different aspects of the power generation expansion planning problem and the need for solution methods that utilize the model characteristics. A general conclusion from this review of the three models/methods is that there is no one "best" model or a "best" solution approach. Considerations of problem granularity level impact the complexity of the resultant model. Modeling complexity and limitations of algorithms and computing power mean that one over-arching model or a solution method is not a practical approach to this problem. We also identify here further research areas for extending the proposed models. [ABSTRACT FROM AUTHOR]

Cloud computing is one of the most recently emerged technology which offers resource on demand facility. Cloud computing is an information technology deployment model that involves entrusting data to information systems managed by external parties on remote servers or "in the cloud." In cloud, resources, in terms of infrastructure, applications and data are deployed via the internet as a distributed service by service provider, which make use of applications without installations. This technology allows much more efficient computation by centralizing storage, memory, and processing as well as band width. However, despite the surge in activity and interest, there are significant, persistent concerns about confidentiality and security of data in cloud computing that prevented business organizations from fully accepting cloud platforms. In this paper, cloud computing security issues have been addressed. In addition, and equally important, an algorithm has been proposed that can be used between service providers and data owners to safeguard multi-way authentications and tighten access control for sensitive data in both public and private. [ABSTRACT FROM AUTHOR]

Hybrid Flowshop Scheduling (HFS) is one of the most comprehensively studied in scheduling problems. HFS has m-machines in each stage so that it is more complicated that regular flowshop scheduling. There are several production systems using HFS, such as steel manufacture, textile, and paper industry [8] [15]. Makespan and total tardiness are two important goals in scheduling to make the scheduling more efficient and to fulfill customer demands. HFS is also NP Hard problems, because the bigger problem will need more time for computation to solve the problem. This research offers simulated annealing algorithm to solve this problem to get minimum makespan and total tardiness. Simulated annealing had used for scheduling problems before and this algorithms gives the good performance. Simulated annealing has the strength to get out of local optima by receiving the worse solution. In this research, this algorithm will be modified by increasing the temperature once when temperature had reached very small value. From the experiment done, modified simulated annealing gives the better performance than regular simulated annealing, especially in the bigger problem. [ABSTRACT FROM AUTHOR]

Milk collection remains a challenging problem for dairy processors around the world. Previous work on milk collection network has not considered solving the location and routing problem simultaneously. The integrated location routing literature on the other hand remains silent on its application to the milk collection problem. We bridge this gap by considering the integrated location routing problem in the context of a milk collection network. The goal of this paper is to analyze the effect of simultaneously solving the location and routing problem vis-à-vis solving location first and routing second on the optimal solution and computational time. We develop a math formulation for the milk collection network location routing problem (MCNLRP) and for the milk collection network vehicle fleet mix problem (MCNVFMP). We then develop an algorithm for solving MCNLRP and compare the results of the algorithm with the optimal solution for various instances of the problem. [ABSTRACT FROM AUTHOR]