All projects require a structure to meet project requirements and achieve established goals. This framework is called project management. Therefore, project management plays an important role in national development and economic growth. Project management includes various knowledge areas such as project integration management, project scope management, project schedule management, etc. The article focuses on the resource-constrained project scheduling known as problem so- called the resource-constrained project scheduling problem (RCPSP). The RCPSP is a part of schedule management. The standard RCPSP has two important constraints, resource constraints and precedence relationships of activities during project scheduling. The objective of the problem is to optimize and minimize the project duration, subject to the above constraints. In this paper, we develop a convolutional neural network approach to solve the standard single mode RCPSP. The advantage of this algorithm over conventional methods such as metaheuristics is that it does not need to generate many solutions or populations. In this paper, the serial schedule generation scheme (SSGS) is used to schedule the project activities using an evolved convolutional neural network (CNN) as a tool to select an appropriate priority rule to filter out a candidate activity. The evolved CNN learns according to the eight project parameters, namely network complexity, resource factor, resource strength, average work per activity, etc. The above parameters are the inputs of the network and are recalculated at each step of the project planning. Moreover, the developed network has priority rules which are the outputs of the developed neural network. Therefore, after the learning process, the network can automatically select an appropriate priority rule to filter an activity from the eligible activities. In this way, the algorithm is able to schedule all project activities according to the given project constraints. Finally, the performance of the Convolutional Neural Network (CNN) approach is investigated using standard benchmark problems from PSPLIB in comparison to the MLFNN approach and standard metaheuristics.

Project management has a fundamental role in national development, industrial development, and economic growth. Schedule management is also one of the knowledge areas of project management, which includes the processes employed to manage the timely completion of the project. This paper deals with the Resource-Constrained Project Scheduling Problem (RCPSP), which is a part of schedule management. The objective of the problem is to optimize and minimize the project duration while constraining the resource quantities during project scheduling. There are two important constraints in this problem, namely resource constraints and precedence relationships of activities during project scheduling. Many methods such as exact, heuristic, and meta-heuristic have been developed by researchers to solve the problem, but there is a lack of investigation of the problem using methods such as neural networks and machine learning. In this article, we develop a multi-layer feed-forward neural network (MLFNN) to solve the standard single- mode RCPSP. The advantage of this method over evolutionary methods or metaheuristics is that it is not necessary to generate numerous solutions or populations. The developed MLFNN learns based on eight project parameters, namely network complexity, resource factor, resource strength, average work per activity, percentage of remaining work, etc., which are calculated at each step of project scheduling, and identified priority rules, which are the outputs of the developed neural network. Therefore, after the learning process, the network can automatically select an appropriate priority rule to filter out an unscheduled activity from the list of eligible activities and schedule all activities of the project according to the given project constraints. Finally, we investigate the performance of the presented approach using the standard benchmark problems from PSPLIB.

Gas Metal Arc Welding (GMAW) is an extensively implemented arc welding process through the control of input process parameters and the metal from the filler wire. Despite its popular use in various industries, the complex interrelationship between the actual bead and the varying welding parameters makes it challenging to predict appropriate bead geometries via mathematical modeling in a continually changing welding process. In this study, the Regression Learner App was used to compare the performance of supervised Machine Learning (ML) predictive models comprising the Linear Regression (LR), Regression Tree (RT), Support Vector Machine (SVM), Ensembles of Tree (ET), Gaussian Process Regression (GPR), and Artificial Neural Network (ANN) using GMAW dataset. The dataset was scaled and normalized at a range of -1 to +1 to facilitate the visualization of the variation effect. The wire feed speed, voltage, weld velocity, unmelted wire length, and melted wire volume were considered as the input parameters to predict the bead geometry. In addition, the five-fold cross-validation was employed to avoid overfitting and poor generalization. Finally, statistical indicators, namely the Coefficient of Determination (R²), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE), were performed on all developed models to evaluate their performance. Thus, the fine tree and ANN models achieved the highest prediction accuracies of 88–91%, signifying their potential use in future research. In short, the present study demonstrated the performance of various supervised ML algorithms for bead geometry prediction, which would assist the selection of appropriately supervised ML algorithms in future studies.

Cracks are one of the main causes of structural failure and they develop in the structures due to various reasons such as fatigue, temperature variation, excessive load, cyclic load, environmental effects, impact loading etc. Thus, structural health monitoring is necessary to avoid risks, damages and failures. So, in order to avoid an extensive failure or accident, the early prognosis of crack in structures is necessary. Visual inspection and some non-destructive testing (NDT) methods for detection of crack are difficult as it requires time, expenses and are quite inefficient. So the alternative methods are motivated to be developed. In this study, vibration analysis, finite element analysis (FEA) and an alternative way which is artificial neural network (ANN) is used to predict, detect and identify the damages in structures. It is found that the theoretical, experimental, finite element analysis and artificial neural network have good accuracy in predicting the crack characteristics.

The aim of this paper is to propose a combined approach for studying entrepreneurial opportuni-ties based on effectual variables. The proposed mixed approach is carried out in three phases. First, entrepreneur’s effectuation variables are selected via a cognitive map and MICMAC meth-od. Second, a Neural Network (ANN)-based model is performed to highlight the emergence of potential entrepreneur’s conception which rely on effectuation key variables with survival and performance. Finally, ANN model is applied based on effectuation variables. Indeed, many con-firmations and interesting findings have been concluded. The results of the proposed approach are essential to understand Tunisian entrepreneur’s thinking and acting in entrepreneurship pro-cess. We make enrichments to the way of theorizing and practicing entrepreneurship, avoiding the idea of mythic entrepreneur.

This research designs information technology (IT) adoption in service system. It determines the role of IT in determining the performance of service delivery processes. It addresses the concept of IT adoption and discusses the design of the key parameters/elements. Based on a detailed questionnaire survey along with case studies, it outlines how IT can be implemented successful-ly. Its sole purpose is to help establish whether or not IT adoption improves service quality and firm performance.

Abrasive Water Jet Machining is one of the novel nontraditional cutting processes found diverse applications in machining different kinds of difficult-to-machine materials. Process parameters play an important role in finding the economics of machining process at good quality. This research focused on the predictive models for explaining the functional relationship between input and output parameters of AWJ machining process. No single set of parametric combination of machining variables can suggest the better responses concurrently, due to its conflicting nature. Hence, an approach of Multi-objective has been attempted for the best combination of process parameters by modelling AWJM process using of ANN. It served a set of optimal process parameters to AWJ machining process, which shows a development with an enhanced productivity. Wide set of trail experiments have been considered with a broader range of machining parameters for modelling and, then, for validating. The model is capable of predicting optimized responses.

Laser direct structuring (LDS) is very important step in the MID process and it is a complex process due to different parameters, which influence on this process and its final product. Therefore, it is very important to use a reliable model to predict, analyze and control the performance of the (LDS) process and the quality of the final product. In this work we develop mathematical models by using Artificial Neural Network (ANN) and Response Surface Methodology (RSM) to study this process. The proposed models are used to study the effect of the LDS parameters on the groove dimensions (width and depth), lap dimensions (groove lap width and height) and finally the heat effective zone (interaction width), which are important to determine the line width/space in the MID products and the metallization profile after the metallization step. We also study the relationship between the LDS parameters and the surface roughness which is very important factor for the adhesion strength of MID structures. Moreover these models capable of finding a set of optimum LDS parameters that provide the required micro-channel dimensions with the best or the suitable surface roughness. A set of experimental tests are carried out to validate the developed ANN and the RSM models. It has been found that the predicted values for the proposal ANN and RSM models were closer to the experimental values, and the overall average absolute percentage errors were 4.02 % and 6.52%, respectively. Finally, it has been found that, the developed ANN model could be used to predict the response of the LDS process more accurately than RSM model.

In this paper, an attempt has been made to develop a mathematical model in order to study the relationship between laser cutting parameters such as laser power, cutting speed, assist gas pressure and focus position, and kerf taper angle obtained in CO2 laser cutting of AISI 304 stainless steel. To this aim, a single hidden layer artificial neural network (ANN) trained with gradient descent with momentum algorithm was used. To obtain an experimental database for the ANN training, laser cutting experiment was planned as per Taguchi’s L27 orthogonal array with three levels for each of the cutting parameters. Statistically assessed as adequate, ANN model was then used to investigate the effect of the laser cutting parameters on the kerf taper angle by generating 2D and 3D plots. It was observed that the kerf taper angle was highly sensitive to the selected laser cutting parameters, as well as their interactions. In addition to modeling, by applying the Monte Carlo method on the developed kerf taper angle ANN model, the near optimal laser cutting parameter settings, which minimize kerf taper angle, were determined.

Measuring different risk factors such as credit risk in banking industry has been an interesting area of studies. The artificial neural network is a nonparametric method developed to succeed for measuring credit risk and this method is applied to measure the credit risk. This research’s neural network follows back propagation paradigm, which enables it to use historical data for predicting future values with very good out of sample fitting. Macroeconomic variables including GDP, exchange rate, inflation rate, stock price index, and M2 are used to forecast credit risk for two Iranian banks; namely Saderat and Sarmayeh over the period 2007-2011. Research data are being tested for ADF and Causality Granger tests before entering the ANN to achieve the best lag structure for the research model. MSE and R values for the developed ANN in this research respectively are 86×?10?^(-4) and 0.9885, respectively. The results showed that ANN was able to predict banks’ credit risk with low error. Sensibility analyses which has accomplished on this research’s ANN corroborates that M2 has the highest effect on the ANN’s credit risk and should be considered as an additional leading indicator by Iran’s banking authorities. These matters confirm validation of macroeconomic notions in Iran’s credit systematic risk.