Search

This research paper proposes an intelligent forecasting system for the online retail of fashion articles. It takes data from an online fashion retailer, comprising article information, customer information, and corresponding transactions between them. The proposed forecasting model leverages ML and deep learning techniques to be applied to the merged dataset with features necessary for the prediction of sales. Python language is used for executing the research work. By employing supervised ML regression algorithms and AI algorithms on labelled data, the aim is to select the best-fit model based on performance metrics. After comparing the results of RMSE score metrics with other algorithms, the algorithm giving the better result is chosen as the model for prediction. Conclusively the best fit model is further deployed to be integrated in a web-based application using Flask- a micro framework for Python. [ABSTRACT FROM AUTHOR]

The increase in renewable generation in distribution networks is changing the paradigm for system operators. This paper proposes a methodology to deal with the power flow critical events in the electric distribution grid due to a high penetration of distributed generation and energy consumption. The formulation provides energy peak shaving of the overall system with the integration of Energy Storage Systems (ESS) and a dynamic reconfiguration of Remote Controllable Switches (RCS) to avoid possible power flow violations while taking into account the operational costs of the assets. A multi-objective optimisation is posed to schedule the day-ahead operation of ESS and RCS with a new robust approach to take into consideration the uncertainty of the distributed generation and the load demand forecasts. The non-convex and non-linear mixed-integer problem is solved with a novel combination of the Particle Swarm Optimisation and Genetic Algorithm meta-heuristic algorithms on a 33-bus system. [ABSTRACT FROM AUTHOR]

The accurate operation and planning of power plants depend heavily on the ability of utilities to estimate load demand, which also enables them to allocate resources more efficiently and guarantee a steady flow of electricity. Modern models have limited relevance in real-world situations and demonstrate severe non-linearity in load information from forecasts that are now accessible. But for practical use, the field of energy forecasting needs more flexibility, enhanced prediction accuracy, and resilience. This study proposes a brand-new method for anticipating short-term load demand that has been uniquely developed for Bangladesh. An extensive dataset made up of historical load demand data, weather factors, and Bangladesh-specific calendar data is used to train and test the model that is suggested. The dataset spans six years of load demand data, covering several seasonal trends. The proposed model uses a parallel CNN-BiLSTM architecture. In this architecture, the BiLSTM module keeps track of temporal relationships and learns long-term trends, while the CNN module pulls out spatial features from the merged dataset. The parallel nature of the model enables it to capture the intricate correlations visible in the load demand time series data. The model also makes use of parallel processing's advantages, which boost the effectiveness of computation. Based on the results of experiments, the proposed CNN-BiLSTM hybrid computational neural network system is better at making accurate predictions than baseline models and traditional forecasting methods. The model successfully captures short-term load demand changes, enabling providers to build facilities, balance load, and schedule resources accordingly. The Mean Absolute Percentage Error (MAPE) and R-squared score, which surpass the existing state-of-the-art models, were taken into consideration for evaluation. [ABSTRACT FROM AUTHOR]

The service sector has been seeing growth over the past few years, and this trend is anticipated to continue. The ability to serve customers is one of the key characteristics of a business. The most important thing is to use the available resource efficiently. To do this, businesses are utilizing cutting-edge technology. An intelligent sequence model is furthermore needed for the prediction of taxi demand when coupled with a time series forecasting issue. We discuss the challenges in predicting taxi demand, such as data sparsity,and propose solutions to address these challenges. Our outcome shows that our model outperforms existing methods in terms of prediction accuracy and computational efficiency. [ABSTRACT FROM AUTHOR]

To offer the patients' medical care and successful outcomes, a successful hospital management is required. Traditional hospital management systems, which typically rely on manual procedures, are unable to handle the vast amount of data that is generated in the field of health in an effective manner. The statistical management method for healthcare systems that we discuss in this paper is based on machine learning. A significant amount of hospital-related data, including patient records, medical procedures, staffing levels, and resource distribution, are evaluated using the methodology, which makes use of automatic learning capabilities. The paper's objective is to create predictive models that, through patient demand projection, resource optimization, and resource improvement, assist hospital management in making better decisions. To correctly forecast future patient loads, bed occupancy rates, and staffing requirements, analyze previous data to identify patterns and trends. These predictions can assist hospital administrators in efficiently allocating resources, cutting down on patient wait times, and raising patient satisfaction. Additionally, our approach gives hospital administrators dashboards and visualizations that provide crucial performance indicators, merging data-driven decision-making, so they can monitor and assess the effectiveness of their management strategies. Utilizing automatic learning algorithms ensures the adaptability and scalability of the hospital management system. [ABSTRACT FROM AUTHOR]

The application of Big Data Analytics, sometimes referred to as BDA, is becoming an increasingly crucial part of the field of supply chain management (commonly known as SCM), which is becoming an increasingly important field overall. This is due to the fact that B is flexible enough to be utilized in a diverse range of SCM-related processes, which is the rationale behind the aforementioned remark. This category includes a wide variety of different types of employment, such as researching trends, forecasting demand, and evaluating customer behavior, to name just a few examples. With the help of this literature analysis, our goals are to categorize the predictive Big Data Analytics applications in supply chain demand forecasting, identify the research gaps, and provide some ideas for further investigation. In the tough economic environment of today, firms have used a wide array of targeted marketing tactics in order to maintain or expand their profit margin while simultaneously being one step ahead of their competitors. The usage of forecasting models is a significant component of the practice of precision marketing which aims to better comprehend and satisfy the requirements of the customers it serves. The evaluation of consumption patterns and preferences on the basis of data collected from customers and records of transactions is receiving an increasing amount of focus and attention as a result of this trend. This is done to ensure that product supply chains (SC) are managed in the most effective manner possible. The basic purpose of supply chain management, commonly known as SCM, is to exert command over the flow of products, services, and information as it goes from the locations where it was made all the way to the individuals who will ultimately buy it. The presumption that capacity, demand, and cost are all deterministic is the foundation of the great majority of research on the application of SCM to problems that occur in the real world. In point of fact, the presence of uncertainty is made worse by a variety of issues, some of which include but are not limited to varying customer demand, management of the supply chain, organizational risks, and lead times. To be more specific, demand ambiguity has a significant impact on SC performance and has far-reaching repercussions for the production scheduling, inventory planning, and transportation organization. This is because demand ambiguity makes it difficult to accurately predict future demand. In addition to this, the unpredictability of the market's demand might have a negative impact on the efficiency of SC's operations. As a direct result of this fact, demand forecasting is an essential instrument for the mitigation of chain risks. [ABSTRACT FROM AUTHOR]

This study aims to improve production planning by utilizing forecasting techniques to estimate future demand for a company's products, specifically shrimp crackers. The company often experiences excess or shortage of raw material stock, which can cause disruptions to transportation and product damage if not managed properly. To address this issue, an artificial neural network method was used to forecast sales for 12 consecutive periods from January to December based on sales data from January 2018 to December 2021. The results showed a root mean square error of 0.120, indicating accurate sales predictions. The implication of this study is that accurate sales forecasting can help companies optimize raw material usage and improve production efficiency, ultimately reducing waste and costs. [ABSTRACT FROM AUTHOR]

Accuracy in forecasting is vital to determine a precise result in every decision-making. However, massive spikes or outliers in a data series would interrupt the forecasting process. Hence, the smoothing method was introduced to prevent these factors from influencing forecast points. Throughout this study, 4253HT smoother was used to smooth a data series before Holt-Winter was applied in forecasting. Moreover, a comparison was made between the use of raw data and smoothed data in forecasting. The raw and smoothed data performance was evaluated using Residual Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Akaike Information Criterion (AIC). As a result, the modelling 4253HT data yielded lower RMSE, MAPE and AIC values compared to raw data. The findings also presented that Holt-Winter performed better in forecasting with models that utilized 4253HT smoothed data. Furthermore, the application of 4253HT smoother on electricity data was made, where the outcomes indicated that it is a good model in data representation as the fitted line moves closer to 4253HT smoothed data. The use of smoothed values in forecasting has generated an excellent outcome as the data outliers are eliminated while its volatility is constant. [ABSTRACT FROM AUTHOR]

Demand forecasting is one of the key aspects of operations management, particularly sales forecasting, as it plays a fundamental role in the retailer's resource planning system, which impacts consumer satisfaction. Traditional time series models are widely used in the context of sales forecasting, while machine learning approaches have been playing an important role in recent years. Different approaches should be tested and compared to support practical knowledge. In this work, we compare different prediction methods to forecast diesel fuel sales in a small business in order to choose the most suitable method, in terms of the usual evaluation measures, for presenting an inventory planning for weakly fuel sales. [ABSTRACT FROM AUTHOR]

Changes in demand for oil palm have an impact on the palm oil industry as a supplier of raw material for palm fresh fruit bunches. Various downstream industries require crude palm oil (CPO) as raw material for the food industry. Product quality becomes very important in advance for the foodstuff industry which burns as the main raw material. The purpose of this research is to make a balance between the demand and supply of oil palm stocks so that distribution is always in accordance with demand, in this case, the level of distributors and recipients that can maintain market prices and ensure the quality and quality of palm oil and that has become CPO and identify problems can occur in the process of domestic and export shipments. This study aims to develop a palm oil supply chain model based on Tsukamoto's FIS (Fuzzy Inference System). The problems above were carried out by research in dealing with problems in distribution using the Fuzzy Logic Method in the problem of CPO distribution problems. Analysis was carried out using the Fuzzy Logic approach. desired in terms of quality and so on. [ABSTRACT FROM AUTHOR]

Air temperature rise due to global warming can increase the demand for electricity, because of the more frequent and intensive use of air conditioners and fans. Based on this phenomenon, it is necessary to study the correlation between air temperatures and peak electric demands. This study aims to obtain a correlation between maximum air temperatures and peak demand and then estimate the peak demand of Padang city for the year 2030. The correlation between air temperatures to peak demand is identified by Pearson coefficient. The demand forecasts are calculated using a multivariable linear regression method. Correlation test results show a low correlation between maximum air temperature and peak demand but a strong correlation between population and peak demand. The results of the forecast for 2030 are 422 MW for case 0, where the independent variables are the population and daily maximum air temperature. The peak demand is 421.59 MW for case 1, where the independent variable is only the population. For case 2, where the independent variables are population and daily maximum air temperature according to Paris Agreement, the peak demand is 421.03 MW. The forecast result that excluded the air temperature resulted in 410 KW lower peak demand than the forecast that includes air temperature. The higher air temperatures result in higher peak demand. [ABSTRACT FROM AUTHOR]

The article is devoted to the issues of forecasting market demand for medicines using modern artificial intelligence (AI) technologies. In a dynamically changing medical landscape and increasing consumer variability, accurately and effectively forecasting demand for medical products is critical for pharmaceutical companies and healthcare systems. The article examines methods and approaches that use AI technologies such as machine learning, big data analytics, and fuzzy sets to create accurate and predictive drug demand forecasting models. Modern approaches to the collection, storage and processing of medical data, which play a key role in the forecasting process, are analyzed. The article also provides practical examples of the successful implementation of AI technologies in the pharmaceutical industry and provides recommendations for their implementation to optimize production, logistics and drug inventory management. The article describes the importance of collaboration between pharmaceutical companies and artificial intelligence specialists, and also puts forward prospects for the further development of this field of research in the context of modern challenges and opportunities in the field of healthcare and pharmaceuticals. [ABSTRACT FROM AUTHOR]

In the present article, the solution of the problem for choosing an optimal size range of linear pneumatic modules, widely used in the automation of discrete production processes, is considered. Choosing the optimal size range of translation modules is essential for good economic performance for both manufacturers and users of these devices. For solving the problem, the main parameters of the devices on which the optimization will be performed are selected – load capacity and stroke length, and their values are determined using series of preferred numbers. Module demand is forecast based on market research. A known optimality criterion is used, which represents the total costs of production and operation. On the basis of a known cost model and functional equations, an algorithm for solving the problem is developed, taking into account the presence of constraints on the values of the modules' parameters. The constraints on the values mean that the next element in the size range must have main parameters' values no less than the previous element. With the help of the developed tools, the optimal size ranges are determined for three different ways of defining the modules' demand: determined as a result of primary data collection from users, determined after linear regression of the principal components, and determined after recovery of the missing values. [ABSTRACT FROM AUTHOR]

Forecasting is a method for predicting future events based on time series data. Most forecasting is done for daily, weekly, monthly, even yearly. Forecasting the Air Pollutant Standard Index (ISPU) in DKI Jakarta can be used for future decision making and preventive measures. The method used is the triple exponential smoothing method from Holt-Winter with alpha = 0.4, beta = 0.01, and gamma = 0.05. Forecasting results are tested by measuring the error rate of the MAPE value and producing a MAPE of 30%, so that the Holt-Winter method has an accuracy of 70%. Forecasting results are visualized on a dashboard for use by local authorities. [ABSTRACT FROM AUTHOR]

Purpose. In this article the authors test the hypothesis on the impact of service transformation during the COVID-19 pandemic on the performance of Russian manufacturing companies. Methods. The study uses least squares linear regression models. The proposed models reflect the logic of transforming input resources into outputs, based on the transformation of incoming labor, material resources, government support and servitization that have become relevant for companies during a pandemic. The empirical base of the study is data from Enterprise Surveys; the authors used materials from a survey of Russian enterprises conducted by the World Bank in the mid-2020. Results. Companies faced a significant reduction in consumer demand and were forced to reduce production volumes and the number of their employees. An increase in online service activity of companies is an important predictor of consumer demand growth in the market for manufacturing companies during a pandemic, while carry-out activities and a corresponding transformation of supply chains are crucial only for the service sector. Practical implications. The results of the study can be applied to support plans to develop servitization strategies of the manufacturing business. In particular, companies need to introduce basic servitization transformations. Limitations and further research. The authors use basic and intermediate servitization indicators related to the product distribution system at the end of the production chain. In further research, it would be rational to use the variables of advanced servitization in order to show its effect on performance. [ABSTRACT FROM AUTHOR]

The article presents the scheme of functioning of the information system "Optimization of production and sale of meat products taking into account the variability of prices and demand", describes its mathematical and algorithmic support. The main stages of planning the production and sale of meat products are highlighted, according to which, at the first stage, trend-seasonal models are built and prices for meat products are predicted taking into account seasonality. At the same time, it is proposed to use the Oracle software to isolate the components of the price time series. At the second stage, the specific sales coefficients of processed products are calculated, reflecting the level of demand for certain types of meat products in a certain period. The third stage is devoted to the creation and solution of a parametric model for the production and sale of meat products, taking into account price and demand forecasts. At the fourth stage, the plans for processing and selling products are adjusted taking into account the decisions received. The given optimization algorithm is implemented on the example of an enterprise that processes meat and sells meat products in the Irkutsk region. According to the results obtained, planning the production and marketing of meat products, taking into account demand, will increase the profit from its sale at the enterprise by 31%. It should be noted that in addition to demand, when modeling meat processing and the sale of meat products, it is necessary to take into account seasonal changes in purchase and market prices. The proposed models are applicable for planning the activities of meat industry enterprises in the short and medium term. [ABSTRACT FROM AUTHOR]

The most important requirement of any energy utility is the accurate prediction of their energy demand. They can use the predicted data to manage their load despatch. Even now many of the energy utilities are using semi human based load prediction techniques for their operation. But these techniques are prone to high errors and are getting obsolete. A number of load prediction models have been developed over the decades using various techniques. It extends from mathematical models to data mining. Such model use regression methods, SVMs, ANNs etc for load forecasting. Till few years the power generation was concentrated in generating stations and its stationary nature made the prediction easy. With the introduction of renewable energy sources, distributed generation has grown into a significant portion as compared to fixed generation. In the present situation the solar distributed generation has become a considerable amount and its uncertainty in production poses a serious challenge to load prediction. This study aims at exploring a suitable model for load prediction integrating distributed generation. [ABSTRACT FROM AUTHOR]

India is one of the largest countries of the world with a high population density. Being a developing country, this nation is facing enormous challenges in regulating its air quality. Population growth, vehicular ownership, urbanization, demand of energy and industrialization process are driving the pollution level to an alarming state in Delhi which is the capital of India. Accurate air quality index (AQI) forecasting of Delhi will assist Government in making strategic decisions so as to avoid catastrophic deterioration in the quality of healthy air. Motivated from this, in this paper, for the first time a systematic study is made to assess the effectiveness of four promising univariate statistical forecasting models and twelve machine learning (ML) models in predicting the AQI of Delhi. Four accuracy measures including "mean absolute scaled error (MASE), symmetric mean absolute percentage error (SMAPE), root means square error (RMSE) and mean absolute error (MAE)" are used to assess the performance of statistical and ML models. Extensive statistical analyses on obtained results using Wilcoxon Signed-Rank test illustrate the superiority of multilayer perceptron (MLP) and autoregressive integrated moving average (ARIMA) models than other considered models in predicting the Delhi AQI. [ABSTRACT FROM AUTHOR]

Customer Churn Prediction in Banking Sector is the most research area in past few years. Nowadays there isvarious options available for customer in banking sector, where to put money. It is important to gain Customer satisfaction and to maintain bank-customer satisfaction relationship with customer. Generally, if a client stops mistreatment the services of an organization for an extended amount of your time (which varies looking on the merchandise or services of the company), such a client is taken into account churned. In this research paper, a method is proposed to to anticipate client stir in financial area using Machine Learning Algorithms with PCA. In this study Support Vector Machine (SVM), Random Forest Classifier (RFC), Linear Regression (LR), Principal Component Analysis (PCA), Feature selection technique are used to find most relevant features. The dataset is taken from Kaggle. After preprocessing of dataset, the dataset is cleaned from the noisy data. And the dataset is ready for the training process. Now the algorithms are used to train the model. In this using PCA with SVM, PCA with RFC, PCA with LR, Accuracy is measured using confusion matrix. As a result, applying PCA with SVM gives high accuracy as compared to others models. [ABSTRACT FROM AUTHOR]

Blood is an irreplaceable thing in the human body. If a person lacks blood then a blood transfusion is needed from someone else. The Red Cross is an organization that handles the need for blood bags for the purposes of blood transfusions. The amount of demand for blood bags varies greatly each month so it is necessary to predict how many blood bags are needed. In this study, the demand for blood bags was forecasted using the ARIMA method. Forecasting was carried out on Whole Blood (WD) and Platelet Concentrate (TC) blood components considering the large difference in expiration time between WD with an expiration of 35 days and TC with an expiration of 5 days. Demand data is taken from January 2018 to September 2021 at the Indonesian Red Cross Surabaya. The forecasting results obtained show that the ARIMA model (0,1,3) for WD and the ARIMA model (1,1,3) for TC has the lowest RMSE error rate. From plotting the forecasting data against real data, it can be seen that the model made is adequate to be used as a forecast of blood bag demand. [ABSTRACT FROM AUTHOR]

The implementation of six sigma analysis in the industry era 4.0 have an important role in production management to produce superior products in accordance with quality standards, both on the standards for management or standards based on consumer demand. The output of industrial products is generally in accordance with product specifications so that the distribution of data tends not to be symmetric distribution or not normally distributed. The research aims to apply EWMA transformation to multivariate six sigma analysis for data that is not normally distributed. The weighted method in the EWMA transformation plays a role in changing the univariate data with normal distribution. Based on data of multivariate normal distribution, the T2 Hotelling control chart were designed using the mean and standard deviation parameters based on the original of data observations. Furthermore, based on the controlled multivariate normal distribution data, estimation of measure of multivariate capability index and six sigma level is carried out. [ABSTRACT FROM AUTHOR]

This paper investigates the use of nonparametric regression to forecast natural gas demand. The demand for natural gas is constantly rising to meet people's daily needs. This causes natural gas capacity to dwindle day by day. The fluctuating increase in natural gas prices is one of the consequences of the current imbalance between natural gas production and consumption. This time, nonparametric regression was used to analyze the data. In this case, the nonparametric regression approach technique can be used as an alternative because its use is not tied to global assumptions. Although it does not respond quickly enough to natural gas demand, it can predict natural gas prices well by taking into account various existing factors. As a result, we require a nonparametric regression model that accurately predicts natural gas demand. [ABSTRACT FROM AUTHOR]

As one of the new provinces in Indonesia, West Papua Province needs the infrastructures for supporting its development, including electrical infrastructures. In this paper, system dynamics modeling has been built to investigate the readiness of electrical infrastructure, especially electricity generation, in facing the energy demand by the growth of population and other sectors in West Papua Province. The models have met the absolute mean error (AME) standard of validation. The simulation of the Business as Usual (BAU) model shows that the generation units are supported well in normal condition until 2050. In case no additional installed capacity for the generation units, the existing units may not support the population and energy growth by 2043 and 2047 in.the designed scenarios simulation. The lack of capacity is due to the upwards of energy demand in 2050, about 5012.09 GWh in scenario 1 and 7406.98 GWh in scenario 2, which is significantly increasing from 5012.09 GWh in BAU. [ABSTRACT FROM AUTHOR]

The current era with very advanced technology, most of which use electricity as an energy source, research is needed for the growth of electrical energy needs. One of them is in the city of Banyuwangi which is one of the cities in East Java that is experiencing rapid development in terms of the industry which automatically affects the growth of electrical energy needs. In this case, electrical load forecasting has an important role in the distribution of the electricity network to maintain a balance between demand and supply of electricity, which research has never been done in the city of Banyuwangi. Therefore, an accurate method is needed to predict the electrical load. So in this study using the Autoregressive Integrated Moving Average (ARIMA) method. The ARIMA method is the fully independent variable in making forecasts. The advantage of this method is that it can accept all types of stationary model data. This study takes medium-term load forecasting from January 2015 to December 2019 for the period 2020. From this research, the error value or MAPE will also be obtained. The beginning of data processing from this research is by looking at the time series plot data, stationary data invariance, and stationary data on average. The results of the three aspects indicate that the data is still not stationary. So it is necessary to transform the data. On the average stationery, it is necessary to make a distinction to obtain stationary data. Next, determine the ARIMA model using ACF for order q and PACF for order p. The result is 2 lags that are out of bounds from the partial image and 1 lag that is out of bounds from the autocorrelation image, because differencing is only done 1 time, the possible models are (2,1,1) (2,1,0) (1,1, 1) (1,1,0) (0,1,1) and then testing the parameters. [ABSTRACT FROM AUTHOR]

With the advent of new technology and Industry 4.0, innovative approaches to increase Supply Chain visibility are necessary. Collaborative planning, forecasting, and replenishment (CPFR) can all improve supply chain visibility (SCV). It is expected that synchronizing project planning, forecasting, and determining replenishment rates significantly contributes to supply chain performance. This research used a literature review to synthesize prior research and analyze the effect of CPFR implementation. In this research, we concluded with the state of knowledge gaps in CPFR and the potentials for future research. We contributed by reviewing the recent five years of literature and provide an essential view for research to fill the literature gap. [ABSTRACT FROM AUTHOR]

The study aimed to analyze how strong the relationship between egg demand and linear factors is. Chicken eggs have a contribution of animal protein for the community, so their existence needs to be maintained. The study was conducted in Biringkanaya District, Makassar City which was selected purposively by considering this area has the largest population. Sampling was carried out by proportional random sampling method, in 5 regions that had the largest number of households, so that the number of respondents who were considered capable of representing the total population was 50 respondents. The results showed that the tested variables X and Y variables had a significant correlation, which was supported by the F Change value of 0.001, so the null hypothesis (H0) was rejected and the alternative hypothesis (Ha) was accepted. Meanwhile, there is simultaneous multiple correlations between the independent variables, the price of layer eggs (X1), the price of chicken meat (X2), the price of tempeh (X3), the price of tofu (X4), and income (X5). and the dependent variable demand for layer eggs (Y) has a strong and positive relationship seen from the correlation coefficient (R) of 0.615. [ABSTRACT FROM AUTHOR]

In present global competitive environment, race is not among the organization but in the system of supply chains (SC). A system of supply chain is a link/networks of participants (suppliers, manufacturers, distributors, and retailers) comprising of different stages (echelons) which are directly or indirectly connected together by movement of materials, information, and capitals with a goal to optimize the total cost of the system. In this work, we study the use of stochastic demand-based techniques for assessing/analyzing the performance/functioning of serial structured supply chain networks system. Computing the lead times of make to order and make to stock supply chains. This paper analyses the performance of 4-stage serial structured supply chain under (R, S) system. In particular, we have shown the use of static probability networks for computing the lead time. [ABSTRACT FROM AUTHOR]

Based on the original Grey Model GM(1,1), the forecast accuracy of GM(1,1) have been modified using Fourier Series called FGM(1,1) to predict the demand of air compressor of compressor manufacturing industry. Markov Chain method have also been adopted to determine the optimal inventory with the minimum cost. The first step in this research is to determine the forecast value of demand for July 2020- December 2020 based on time series data in August 2019-June 2020. The calculation results showed that Fourier Series Grey Model (FGM) works better than the original GM(1,1), as evidenced by the better level accuracy of FGM(1,1) which are 9.66% of MAPE and 10.14% of RMSPE, while GM(1,1) gives 18.83% of MAPE and 23.64% of RMSPE. The next step is using time series data in August 2019-June 2020 and FGM forecasting results for inventory planning using Markov Chain Method. The results of the optimal inventory level can reduce the cost by 23%-64% of the company's current inventory cost. [ABSTRACT FROM AUTHOR]

The percentage of electrical energy needs is the largest demand compared to other energy needs such as natural gas, fuel, and coal. This is due to various factors, including population growth, economic growth, industrial development, as well as the rapid development of electricity-based technology in almost every sector, especially in the household, industrial and commercial sectors. This condition has the potential to trigger an electricity crisis, but can be minimized if the required electricity consumption is known. One way to determine electricity demand is to build a predictive model that is accurate and flexible and able to accommodate the complexity of seasonal patterns, both seasonal in months and years as reflected in electricity consumption data. Therefore, in this study, the TBATS model was used to accommodate this. TBATS models will be used which is the development of the exponential smoothing model that can accommodate the occurrence of multiple seasonal patterns, both nested and non-nested, non-integer seasonal periods, and handle the possibility of non-linearity cases because it has a flexible seasonality The results of this study, the TBATS model built has a value of SMAPE by 8.13% has been able to capture fluctuating patterns in seasonal periods. [ABSTRACT FROM AUTHOR]

PT VIS is a company engaged in the automotive industry of electrically powered vehicles. Currently, PT VIS is developing an electric motorcycle product to add to its existing product line. In developing new products, some risks and obstacles cannot be avoided. Therefore, a risk assessment and mitigation actions must be carried out, which can be done using the House of Risk method. Based on the data collection, 17 risk events are relevant to the company. From the data processing that has been done, two risk agents with the most significant Aggregate Risk Potential values were identified, namely the battery with a limited lifetime and poor cooperation between teams. Based on this, preventive actions were modeled to prevent risks. The selected mitigation actions are stock calculation following the demand forecasting carried out, proper battery storage SOPs, more supervision from the head of the department on the team's performance in the production process, more communicative use of language from the head of the department to the staff, market research more profound into the market trends and target, as well as appropriate calculation in demand forecasting within the market research carried out. [ABSTRACT FROM AUTHOR]

In the context of the transition to a new technological order, the issues of determining the strategic goals of innovative development, tactical measures, optimizing the interactions of innovation actors in the context of the networkization of the economy require an updated, scientifically grounded approach based on systematic support for managerial decision-making in order to activate innovative growth in the process of self-organization of industries and complexes. Scientifically substantiated support for decision-making on the problems of innovative and strategic development at the micro and meso-level of the economy will allow not only activating the existing ones, but also creating new "points of innovative growth" that contribute to the active and effective diffusion of innovations in both industry and regional projections. Modeling innovative and strategic development at each level of the planning horizon: from foresight to building interactions between participants in the innovation network, will not only contribute to the adoption of well-grounded, balanced and effective management decisions, but also form such guidelines for innovative development that will minimize costs and risks, while maximizing the creation of demand for innovative products and the activation of innovative activities. [ABSTRACT FROM AUTHOR]

The framework aims at analyzing the telecom customer data and predicting the churn for improving the customer retention rate. The customer information dataset used for training and testing of the proposed prediction model consists of 7043 customers with 21 attributes. The Prediction of Churn is done by adapting Random Forest (RF) and Support Vector Machine (SVM) algorithms. The classification accuracy of the Random Forest classifier is (79%) and SVM is (75%). There is a statistically significant difference among the study groups with a significance value (p<0.05). The experimental results provide evidence that the Random Forest classifier seems to be more suitable for churn prediction with a lesser error rate, better sensitivity, and accuracy. [ABSTRACT FROM AUTHOR]

Urban areas are having an increase in population growth extensively and the need for transport is increasing. Number of available taxi drivers is beginning to become insufficient and we can experience this in our day to day lives. Taxis are sparsely available late at night which makes it dangerous for anyone to be around, People end up having to wait long periods of times under extreme conditions which decreases the taxi services over satisfaction rating, drivers are unaware on where to head after dropping of a passenger and look for the next fare, taxi drivers are unwilling to drive to a comparatively remote location in fears of not finding any fares there and worry about wasting time and fuel. Such cases can be solved if the demand of taxis can be predicted. Given a region, I want to be able to predict the demand so as to inform dispatch systems that can allocate taxis. [ABSTRACT FROM AUTHOR]

A long-term electricity demand forecasting on the West Kalimantan electricity system using the Simple-E program (from Japan) has been researched, with the study period of 2021 - 2050. Projection of electricity demand growth is conducted by using 3 scenarios, namely: Basic Scenario (Business as Usual (BaU)) where this scenario uses the econometric regression analysis methods; Industrialization Scenario where this scenario uses Basic Scenario (BaU) with added in the plan for the development of industries in West Kalimantan; and Prosperity Scenario where it takes into account the electricity consumption targets per capita in National Energy Policy (Kebijakan Energi Nasional, KEN). The result of the Basic Scenario for electricity demand projection in 2021 is predicted to reach around 3.02 TWh and projected to increase to reach around 15 TWh in 2050. Meanwhile, under Industrialization Scenario, the Basic Scenario (BaU) projection results are injected or added with the electricity demand from industrial and business sectors planned to be developed in West Kalimantan. The projection of electricity demand based on the Industrialization Scenario in 2021 is predicted to reach around 3.02 TWh and projected to increase to reach around 18.7 TWh in 2050. The Prosperity Scenario considers the electricity consumption targets per capita stated in the Energy National Policy (KEN). KEN mandates the electricity consumption per capita for Indonesia of 2,500 kWh/capita in 2025 and 7,000 kWh/capita in 2050. Based on this consideration, the Prosperity Scenario projection of electrical energy demand shows that it rises from initially at around 3.86 TWh in 2021 to reach around 30.1 TWh in 2050. Based on that three scenarios, the projection of electrical energy demand using the Prosperity Scenario is higher than the BaU and Industrialization scenarios. Using the Prosperity Scenario, electrical energy demand in 2050 also reaches twice based on BaU Scenario. This result is due to considering the electricity consumption target per capita, which is relatively high following the target in KEN. Under Industrialization Scenario, the injection of electrical energy demand in the industrial sector causes a jump in electrical energy demand in 2025, exceeding the Prosperity Scenario caused by the presence of some industrial zones targeted to start operation in that year. [ABSTRACT FROM AUTHOR]

The objectives of this research is to determine the forecasted demand and the best forecasting method that can be used to predict the demand of frozen tuna at PT Prima Indo Bahari in 2018 and get the best aggregate planning strategy for the company. This research used quantitative method and the type of this research is descriptive research. The demand data from past 3 years is used to forecast the demand using linear regression methods and seasonal index. Forecasted demand is used to design aggregate planning using 3 alternative strategies Chase Strategy, Level Strategy, and Mixed Strategy which is computed manually and using POM-QM for Windows 4. The results of this research show the forecasted demand in 2018 is 1.252.145 Kg and Mixed Strategy is the best strategy that can meet all the demand of frozen tuna at PT Prima Indo Bahari with the lowest cost at Rp 19,651,170,000. [ABSTRACT FROM AUTHOR]

In dealing with the Covid-19 pandemic, the government issued policies to limit economic activity and reduce electricity demand. However, to overcome this, the government has carried out a Covid-19 vaccination program to achieve Herd Immunity to resume their economic activities to increase electricity demand for several sectors. To analyze electricity in Central Java after the vaccination program during Covid-19 then an energy planning model was made using the LEAP software. LEAP requires data in social-economic data, customer data, and energy use intensity per sector. The scenarios used in this study are the no pandemic or business as usual (BAU) scenario and two post-COVID-19 pandemic scenarios with vaccination policies, namely optimistic (OPT) and pessimistic (PES) scenarios. From the results of this research, the projected results of electricity demand in 2030 are 40,691 Gigawatt-Hour (OPT) and 39,924 Gigawatt-Hour (PES). Therefore, it is necessary to consider the consequences of the Covid-19 Pandemic, where in the basic scenario, 43,773 Gigawatt-Hour, is obtained. [ABSTRACT FROM AUTHOR]

Reasoning with the Sugeno method is almost the same as Mamdani reasoning, it's just that the system output is not a fuzzy set but a constant or linear equation. In this Study the data is obtained from Micro Enterprises which are engaged in selling chicken eggs and distributing them to consumers. based on sales, demand, and inventory data obtained from March to May 2021 and after using the Takagi-Sugeno method, the Mean Absolute Percentage Error (MAPE) of 18.13% was obtained. The results of the work with the output of the matlab tools show the MAPE result of 15.81, which means that the prediction accuracy reaches 84.19%, which means that the forecasting interpretation is good. [ABSTRACT FROM AUTHOR]

The increase in population that continues to increase every year and the national economic growth impacts the increase in chili demand, both from households, processing and food industries, and export needs. The price also influences the demand for chili products in the market. When chili production is abundant, the price of chili will be very low and demand will increase, but on the contrary, when chili production decreases, the price of chili will be high and demand will decrease. Therefore, this study aims to analyze the factors that influence Indonesian chili demand and determine the elasticity of chili demand in Indonesia. The data used in this study is data from 2005 to 2019. The analytical method used is multiple linear regression analysis. The results showed that the independent variables that significantly affected the Indonesian chili demand are income per capita, garlic price, and chili export price. The price elasticity of 0.134 indicates that the demand for red chili is inelastic. Shallots and garlic are complementary goods with elasticities of -0.285 and -0.060. The income per capita elasticity is 3.115, indicating that the percentage of red chili demand will increase if income increases. [ABSTRACT FROM AUTHOR]

The accuracy of electricity load demand forecasting is essential for avoiding energy waste and overuse. Hence, this paper aims to model the forecast electricity load demand by combining Empirical Mode Decomposition (EMD) with Group Method of Data Handling (GMDH) model. The proposed methodology works in three steps: it decomposes the original load data series into several Intrinsic Model Functions (IMFs) and one residual component, enables individual forecasting of each IMF and the residual using the GMDH model by using the Partial Autocorrelation Function (PACF) as the input variable, and aggregates all the forecasted values to yield the final prediction for electricity load demand. To compare the performance, another model is considered namely the combination of EMD with the Artificial Neural Network (EMD-ANN). The empirical result from the performance evaluation concluded that EMD-GMDH outperforms the EMD-ANN as well as the GMDH model without decomposing the time series. [ABSTRACT FROM AUTHOR]

The purpose of this study is to solve the problem of production planning to minimize overstock with forecasting methods where the results are used to form a Goal Programming model and solved by the Weighted method. Overstock problems can be solved by using the demand forecasting method by comparing the MAD, MSE, and MFE values of four forecasting methods including the Double Moving Average, Single Exponential Smoothing, Double Exponential Smoothing, and Linear Trend method. Forecasting results we obtained are then used to form the Goal Programming model. Another goal to be achieved is to reduce inventory costs and also get profit from sales so that the Weighted method is used here because each target has a different importance weight. The amount of the estimated cost of inventory due to forecast in 2020 will decrease from Rp. 8,130,505,919 to around Rp. 4,589,859,000. Sales profit is obtained with an income of Rp. 379,942,200. [ABSTRACT FROM AUTHOR]

This study identifies the factors that contribute to the students' English language proficiency among Malaysian Institute of Information Technology (MIIT) undergraduates. It further explores the impacts of the undergraduates' language proficiency towards their academic performance. A quantitative and non-experimental approach were used for data collection purposes where data were collected through online questionnaire. There were 116 participants from Universiti Kuala Lumpur (UniKL) MIIT undergraduates involved in this study. Data analyses were made using inferential and descriptive statistical analyses. Statistical analyses revealed that the undergraduates' academic achievement was influenced by their attitude towards the use of English and their high school English performance. In contrast, the findings indicated that the undergraduates' residency did not contribute to their English language proficiency. It was also indicated that there was a significant difference between the undergraduates' level of English Language Proficiency based on their academic achievement. In brief, the findings of this study provide useful insights on how undergraduates' English language proficiency is influenced by many external factors that eventually affect their academic performance. It is crucial to be proficient in English language so that not only students are able to perform excellently in school but also in tertiary institutions. After all, graduates who are proficient in English will be in demand specially to cater the need to work in the English medium environment. Thus, this research paper will provide some improvements and recommendations in assisting the tertiary institutions to produce more in demand graduates who are able to compete at the international level. [ABSTRACT FROM AUTHOR]

PT Trakindo Utama is Indonesia's official dealer for heavy equipment, machinery, generators, forklifts and Cat parts from Caterpillar. The current problem is the company has used the min-max method in spare part inventory control but stock out often occurs in spare part inventory. Therefore, this research aimed to analyze the spare parts inventory to find out materials that need more supervision in order to minimize the possibility that can harm the company, along with inventory planning to obtain the optimal order quantity to minimize the total inventory cost. This research used ABC-FSN analysis of 167 types of 320D Excavator spare parts. ABC analysis is the classification of inventories based on the money value of the goods to the total annual investment of the stored goods, consisting of class A which is goods that give a high value, class B which is goods that give a medium value and class C which is goods that give a low value. FSN analysis is an inventory classification that groups materials into three categories, namely Fast-moving (F), Slow-moving (S) and Non-Moving (N) (deadstock). Then do the inventory planning for materials that are in the AF category. In inventory planning, demand forecasting and determination of optimal order quantities are made using two lot sizing methods, namely Part Period Balancing (PPB) and Wagner Within Algorithm (AWW).Results from ABC-FSN showed that there are 8 types of spare parts categorized as AF which are materials with high investment levels and fast movements. The lot sizing method gives the result that the PPB method produces a minimum total inventory cost compared to the AWW method. [ABSTRACT FROM AUTHOR]

The paper proposes energy prediction techniques on Electric energy consumption from existing electricity supply using machine learning. Machine learning and its techniques are vastly used in factories for maintenance, in industries for optimization and to calculate the fault tolerance to produce an efficient monitoring system. This paper proposes the best model to monitor the usage of power for predicting the energy Raspberry Pi 3Model B+ is the core controller used for this energy prediction and modelling. AC-current, AC-voltage and power are the input parameters measured using current sensor (ACS712) and AC-Voltage sensor (ZMPT101B). MCP3008 ADC is interfaced with the controller to read analogue current and voltage values. Linear regression and logistic regression are used for both prediction and optimization. The two algorithms are implemented on the Spyder IDE which is installed in the Raspberry pi 3Model B+ controller. After reading the input parameter it predicts the energy consumption and displays whether it is high or low for that day. Experimental Results suggests that the linear regression model is the best among the two because the accuracy metric values viz., Mean Absolute Error, R2 score, Execution time, Residual Sum of Square are comparatively less when compared with logistic regression model. [ABSTRACT FROM AUTHOR]

The article provides information about the use of artificial neural networks and discusses the prediction of energy consumption using an artificial neural network. The article analyzes energy networks and the possibility of using artificial neural networks. The article describes the options for developing a mathematical model. A huge amount of research has been conducted in the field of organization and project management. The article asserts that artificial neural networks improve any process of activity, and it would be reasonable to use it in construction work related to the forecasting of cost, labor productivity and equipment, risks and labor safety. The article provides an example of the possible potential for savings with the use of neural networks in civil engineering facilities. [ABSTRACT FROM AUTHOR]

The article analyzes the state of electricity generation in enterprises of the energy system in 2005-2020 and develops proposals to improve electricity supply. Recommendations on the use of renewable energy sources have been developed for existing electricity generating companies. The mathematical model of this process was analyzed and on the basis of the target to be achieved in 2030, the country's electricity supply by 2030 was determined. [ABSTRACT FROM AUTHOR]

Accurate electricity consumption forecast has primary importance in the energy planning of the developing countries. During the last decade several new techniques are being used for electricity consumption planning to accurately predict the future electricity consumption needs. But still they lag in accurate electricity prediction. To address this problem and to accurately predict the electricity consumption in future, Enhanced LSTM architecture has been proposed named Enhanced LSTM (E-LSTM). In this proposed E-LSTM, feature extraction and prediction has been carried out. A new layer named veracious layer has been added to the LSTM to improve the prediction accuracy of the model. Result shows that the proposed technique is outperformed the existing techniques in terms of prediction accuracy and training loss. [ABSTRACT FROM AUTHOR]

The study objective was to investigate whether the Push factor or Pull factor contributing to overseas business investment. For the study data were collected from 108. The sample considered for the study were companies that recently internationalized their business a year before. The sample was selected from India's export directory. The samples were chosen using a simple random sampling method. The data were collected using a structured questionnaire through an online survey. Through the result obtained it could be understood that Competition in the domestic market, Less Profit due to competition, and Less demand in the domestic market are the major push factor for companies to invest overseas. Further, it is found that Scope for High Demand in the international market, Conversion of National level company to International Brand, Market Size and High-Profit Marginis the major pull factor for overseas investment. Overall, it was understood that; Scope for High Demand in the international market, Competition in the Domestic market, less profit due to competition in domestic are the major reason for overseas investment and internationalization of business. Hereby, it can be concluded that it is a push factor that is majorly contributing to overseas investment. [ABSTRACT FROM AUTHOR]

R&R is a company located in Menganti, Gresik, East Java that sells poultry with a short life cycle. The purchase price of poultry from suppliers and customer demand fluctuates daily. The objectives of this research are suggesting daily demand forecasting, predicting selling price, and determining the optimal single order quantity (SOQ) to maximize profit for the company. The results of the study showed that the SOQ method could maximize the profit by considering the results of daily demand forecast, selling price and purchase price prediction to compare with the real condition. During the planning horizon, the increase of total profit was 60.23% and 6.39% in total revenue. There is an application that was developed to support determination of optimal quantity of poultry by the company. [ABSTRACT FROM AUTHOR]

The article analyzes the current methodological approaches to the analysis of the online labour market and identification of the demand-supply imbalance. The article also shows the limits for the application of these approaches for the labour market in Belarus. The authors have developed and tested a methodological approach to the collecting a database on vacancies from online job portals, classification and processing of the collected data. The research helped to determine the demand for personnel in textile industry and its matching to the structure of CVs posted by applicants. Thus, the data collected from online job portals help educational institutions develop more precise enrollment plans which ensures timely training of personnel that is required by the industry. In the methodological aspect the article illustrates how the use of Big Data improves analysis and forecast of the demand in the labour market. [ABSTRACT FROM AUTHOR]

There is an increment within the power charge of Universiti Tun Hussein Onn Malaysia (UTHM) lately. Thus, there is an ought to anticipate UTHM power utilization precisely to arrange for future vitality demand and utility saving decisions. Past researches on UTHM power utilization prediction have been carried out utilizing time series models, multiple linear regression, 1st-order fuzzy time series (FTS) and seasonal Autoregressive Integrated Moving Average (SARIMA). The 1st-order FTS yields the best accuracy among these four methods. Past forecasting studies showed higher order FTS can yield better accuracy. Thus, the 3rd -order FTS with trapezoidal membership function and 2009 to 2018 of monthly electricity consumption data were utilized to predict 2019 monthly electricity usage. The process of the 3rd -order FTS using trapezoidal membership function was shown in detail using January data. The process was repeated with other months. It was proved that 3rd -order FTS performs the best in overall data, but 2nd-order FTS predicts the future the best. [ABSTRACT FROM AUTHOR]