Forecast Product Demand Research Articles

Optimization of Fuzzy Economic Order Quantity using Machine Learning and Weka

Inventory management is a crucial need for businesses because it requires a significant financial and human resource investment. Machine Learning models are used by e-commerce huge companies to manage their inventory to the demand for specific products. Businesses can benefit from inventory management as a service to increase sales and forecast product demand. Inventory Management might be revolutionised by AI and ML technologies, which would allow businesses to automate repetitive processes, make data-driven choices, and instantly optimise inventory levels. By analysing vast amounts of both historical and current data, these technologies can spot inclination, forecast, and recommend, all of which increase inventory accuracy and lower the chance of stockouts or overstocking. The most efficient solution is determined using the Kuhn-Tucker Method for Nonlinear Programming, which has an impact on the monthly cost. The proposed model employs the Signed distance approach for defuzzification and the Pentagonal fuzzy number to find the lowest cost and met the demand. To evaluate the inventory concept, purchased a dataset and calculated the profit and non profit using apriori algorithm and use the Weka software to analyse the average order quantity and total cost.

Modeling e-commerce retailer demand using Stacking Ensemble and K-means clustering patterns.

Demand forecasting plays a crucial role in e-commerce. By accurately forecasting product demand, e-commerce companies can better manage inventory levels and decrease excess or out-of-stock situations, thereby reducing inventory costs and increasing customer satisfaction. To address the issue of the traditionally low accuracy in market demand forecasting by practitioners who relied on manual experience, this paper proposed a demand forecasting model for e-commerce retailers based on a Stacking ensemble model. The paper successfully modeled and analyzed past user behavior data from JD.com platform, and the model achieved automated forecasting of future demand based on historical data. Upon testing, the proposed Stacking ensemble model demonstrated RMSE (Root Mean Square Error) is 256.34 and 1-WMAPE is 0.916 close to 1, indicating that the model performed well and was capable of accurately predicting the demand for these products. Additionally, this paper also employed a K-means clustering model based on a cost function to classify products into five distinct categories according to their demand characteristic indicators and product attribute indicators.

Perencanaan Produksi UMKM dengan Menggunakan Pendekatan Metode Transportasi

Micro, Small and Medium Enterprises (MSMEs) are an important part of the Indonesian economy. Many micro, small and medium enterprises (MSMEs) are making new innovations in the fields of handicrafts, food and drinks, among others. The development of MSMEs faces demands that change every month. To meet this demand, they must carry out optimal production planning. Production planning and control can be used as a reference for determining production quantities and forecasting product demand. By carrying out good production planning and control, MSMEs can reduce or anticipate risks, thereby increasing their profits. Coconut jelly MSMEs are one of the businesses that are also affected by poor production planning. On average, in 2023 there will be an excess production of 150 pieces per month. For small and medium businesses (MSMEs), fluctuations of this magnitude are very disruptive to company finances. Effective production planning should be able to optimize the use of all organizational resources, including raw materials, materials, labor and finance. This article carries out production planning using a transportation method approach. Initial planning is carried out following a pattern of forecasting the number of requests in the future. Forecasting is carried out using 3 methods, namely single moving average, double exponential smoothing and trend analysis. Of these three methods, the one with the smallest Mean Square Error (MSE) value will be selected. Production planning will be outlined in the Master Production Schedule (JIP). Determination of JIP is based on the results of production planning through 2 alternatives, namely permanent and changing workforce. From the results of the transportation method approach, it was found that the minimum production cost was found in the variable labor alternative, namely IDR 178,326,000, there was a cost difference from the fixed labor alternative of IDR 1,451,200. The approach method used in this article still needs a lot of improvement, especially in terms of determining the cost of each production component which can be more dynamic. The approach that can be taken can use heuristic methods such as genetic algorithms. So, it is hoped that the results obtained can be closer to reality.

Revenue management: changes in the analytical landscape of the enterprise through improving the organization of accounting and analysis

Income management is an important concept in the system of managing the financial and economic activities of an enterprise. A solid revenue management strategy requires the use of analytics and available data to forecast product demand, make strategic pricing decisions, and maximize revenue.
 The article examines the dynamics of the income structure of enterprises and presents a sectoral analysis of companies according to the sectoral distribution of the market environment. The main trends are established and the main problems of the company's income management arising in conditions of uncertainty and risks are outlined. The paper presents the role of income in the management system of the enterprise, which directly affects the efficiency of the process of managing the economic activity of the enterprise.
 In the course of the study, ways of improving the formation and use of accounting and analytical information about the company's income for management purposes were proposed. The scheme of the organization of income analysis for the formation of analytical information is presented. Reasonable importance of management reporting, which is intended to provide understanding, clarity and direction of management activities. When considering the content of the management report, we recommend taking into account visualization forms and design layouts. The KPI dashboard is a visual tool that provides a centralized view of various data sources for effective tracking and optimization of goals. With the help of indicators and KPIs, companies can get an overview of their performance at an operational or strategic level.

RELAYOUT LANTAI PRODUKSI SPRINGBED MENGGUNAKAN METODE SLP, CORELAP SERTA SIMULASI PROMODEL, DAN FLEXSIM

PT. Alpha Jaya Manunggal Mandiri is manufacturing company engaged in furniture sector. There are several problems with old layout such high material handling costs, long distances between work stations, and alternating flow. Therefore, relayout and lean manufacturing approach were carried out. Alternative layouts made using SLP method, and CORELAP. The initial steps in making alternative layouts are measuring floor area, making production flows, and measuring production standard times. Next step is forecasting product demand for next 10 years for routing sheet. Routing sheet used to determine number of machines needed to meetproduct demand. Alternative layout was created using 2 methods. Based on alternative layout, material handling costs and moving moments are calculated to find layout with minimum material handling costs and moving moments moments. The selected layout is alternative layout SLP method cause it reduces material handling costs by 50.12%, moving moments by 17.86%, and distance by 53.42%. Then lean manufacturing approach carried out to increase productivity. A simulation was carried out with Promodel and Flexsim software to simulate new and old layout to find product improvements using new layout. Based on Promodel simulation, there was decrease in the average time in system by 21.82%, decrease in the average time in operation by 22.36%, and increase in total production by 15.29%. Based on Flexsim simulation, there was increase in amount of daily production by 58.52%, and decrease in staytime by 33.47% per product. Through research, it can be concluded that layout can be implemented as new layout on springbed production floor.

Analisis Peramalan Obat Menggunakan Metode Simple Moving Average, Weighted Moving Average, Dan Exponential Smoothing

Forecasting is one of the essential measures in decision making, especially for estimating the number of needs. Some large companies often hire consultants to analyze and advise on forecast product demand. Especially in the health sector, forecasting is needed to meet customer drug demand. In the last two years, the Healthy Prayer Clinic experienced problems in planning the procurement of non-generic drugs due to the COVID-19 pandemic, where the number of drug requests was greater than the number of available drug stocks. This happened because the Healthy Prayer Clinic ordered non-generic drugs without planning. This study aims to determine the demand for non-generic medications in the next period and provide recommendations for the best forecasting method to meet customer needs. The methods used in this research are Simple Moving Average (SMA), Weighted Moving Average (WMA), and Exponential Smoothing (ES). Then test the accuracy of forecasting using Mean Square Error (MSE), Mean Absolute Deviation (MAD), and Mean Absolute Percentage Error (MAPE). The results of the overall forecasting calculation using the Simple Moving Average method of 19.774, then the Weighted Moving Average method of 19.573, and the Exponential Smoothing method of 20.851. The results of the calculation of the best demand forecasting using the Exponential Smoothing (ES) method for 3 months carried out from April 2021 to April 2022 obtained results of 20.851 with forecasting error accuracy using Mean Square Error (MSE), Mean Absolute Deviation (MAD) and Mean Absolute Percentage Error (MAPE) is 1.341.137, respectively; 883; and 0.90%. The best method with the smallest accuracy rate for forecasting errors is the Exponential Smoothing method with forecasting values for the next month, April 2022, amounting to 20.851

How key-enabling technologies’ regimes influence sociotechnical transitions: The impact of artificial intelligence on decarbonization in the steel industry

Key Enabling Technologies (KETs) are pervasive groups of technologies expected to enable innovation. They have been promoted as technologies with tremendous potential for boosting economic growth and sustainability in all sectors of society – a claim whose validity remains underexplored. Building on systems thinking and the Multi-Level Perspective, we develop a novel approach to assess the socio-technical impact of a KET regime on a transitioning sectoral regime. This approach is applied to the case of the Artificial Intelligence (AI) KET impacting the decarbonization of the energy-intensive steel industry. To assess AI's technical impact, we compiled an inventory of AI tools based on reviewing technical scientific articles. Our analysis shows that AI adds technological value to the full range of areas in the steel industry, like predicting process parameters; optimizing operations, scheduling, and electrical energy; and forecasting product demand, quality, and site emissions. Semi-structured interviews were the primary data source to assess AI's broader socio-institutional impact. The results indicate that AI may currently be reinforcing path dependencies of the steel industry, as AI tools are more focused on incremental improvement for existing technologies rather than novel low-carbon technologies. However, AI also offers capabilities to reduce barriers to sustainability innovation, like system integration challenges, flexibility challenges, demand-side barriers, and risk-related barriers. Finally, we reflect on the generalizability of our approach for studying other transitions, and we induce characteristics of the AI-Digital KET regimes. We find that through these characteristics, the AI-Digital KET regime alters existing and creates new system structures (actors, networks, and institutions) within the impacted sector.

A hybrid extreme learning machine model with harris hawks optimisation algorithm: an optimised model for product demand forecasting applications

Accurate and real-time product demand forecasting is the need of the hour in the world of supply chain management. Predicting future product demand from historical sales data is a highly non-linear problem, subject to various external and environmental factors. In this work, we propose an optimised forecasting model - an extreme learning machine (ELM) model coupled with the Harris Hawks optimisation (HHO) algorithm to forecast product demand in an e-commerce company. ELM is preferred over traditional neural networks mainly due to its fast computational speed, which allows efficient demand forecasting in real-time. Our ELM-HHO model performed significantly better than ARIMA models that are commonly used in industries to forecast product demand. The performance of the proposed ELM-HHO model was also compared with traditional ELM, ELM auto-tuned using Bayesian Optimisation (ELM-BO), Gated Recurrent Unit (GRU) based recurrent neural network and Long Short Term Memory (LSTM) recurrent neural network models. Different performance metrics, i.e., Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE) and Mean Percentage Error (MPE) were used for the comparison of the selected models. Horizon forecasting at 3 days and 7 days ahead was also performed using the proposed approach. The results revealed that the proposed approach is superior to traditional product demand forecasting models in terms of prediction accuracy and it can be applied in real-time to predict future product demand based on the previous week’s sales data. In particular, considering RMSE of forecasting, the proposed ELM-HHO model performed 62.73% better than the statistical ARIMA(7,1,0) model, 40.73% better than the neural network based GRU model, 34.05% better than the neural network based LSTM model, 27.16% better than the traditional non-optimised ELM model with 100 hidden nodes and 11.63% better than the ELM-BO model in forecasting product demand for future 3 months. The novelty of the proposed approach lies in the way the fast computational speed of ELMs has been combined with the accuracy gained by tuning hyperparameters using HHO. An increased number of hyperparameters has been optimised in our methodology compared to available models. The majority of approaches to improve the accuracy of ELM so far have only focused on tuning the weights and the biases of the hidden layer. In our hybrid model, we tune the number of hidden nodes, the number of input time lags and even the type of activation function used in the hidden layer in addition to tuning the weights and the biases. This has resulted in a significant increase in accuracy over previous methods. Our work presents an original way of performing product demand forecasting in real-time in industry with highly accurate results which are much better than pre-existing demand forecasting models.

Penerapan Metode Double Exponential Smoothing dan Moving Average pada Peramalan Permintaan Produk Gasket Cap di PT. Nesinak Industries

PT. Nesinak Industries is a company which focuses on the manufacturing process of an electronic component as well as automotive components (vehicle). In business activities, such as production, a strategy is required to survive in competition. Planning and forecasting are a strategy that can be implemented to accomplish these goals. In this study, the data used are previous sealing application data from January 2019 to March 2021. The objective of this study is to forecast product demand over the next period in order to be able to respond to customer demand. Data processing in this study utilize the Brown exponential double smoothing method and the moving average is then determined with the lowest MAPE (Mean Absolute Percentage Error) value to be used for the company’s product demand prediction calculations. The value of taken from Brown's exponential dual smoothing method is the value of with the two lowest error values from 0.1 to 0.9, whose value with the least error value is = 0.8 and = 0.9. In terms of the moving average method, the researchers tested a period of three months and a period of four months. In the MAPE calculation, the results of exponential double smoothing = 0.8 of 26.92 %, exponential double smoothing = 0.9 of 26.22 %, moving average n = 3 of 32.46%, and moving average n = 4 of 34.77%.

Development Of Multi-Criteria Decision Making Model In Packed Beverage Industry Using Global Criterion Method

The packaged beverage industry with various brands has appeared on the market. The number of competitors in the market requires the packaged beverage industry to implement the right strategy to win the competition. One strategy that can be applied is to make savings and maximize the use of company resources. One of the savings that can be done is to minimize production costs, while maximizing the use of resources can be done by maximizing the utility of the machine owned by the company. Efforts to minimize production costs can be contradictory to efforts to maximize machine utility, this is because maximizing machine utility will certainly require additional production costs, even though the company also wants to minimize production costs. This is included in the category of multi-criteria decision-making problems. To facilitate the company's management in achieving these contradictory company goals, this study will develop a multi-criteria decision-making model to assist companies in making optimal decisions regarding these contradictory company goals. The multi-criteria decision-making model developed is a mathematical model consisting of two mutually contradictory objective functions, namely minimizing production costs and maximizing machine utility, as well as problem constraints which are limitations that are owned by the company in optimizing the achievement of these contradictory company goals. To optimize the multi-criteria decision-making model that has been developed, it will use the global criterion method. The global criterion method will produce a compromise solution that has a minimum deviation from the ideal value of each objective function. Using the global criterion method, a production plan that minimizes production costs and maximizes machine utility is obtained. The multi-criteria decision-making model developed will cause the production plan to have a small deviation from the actual demand, because the model development involves forecasting product demand. The developed model will also produce a production plan that will not exceed the warehouse capacity, so that product storage costs can be minimized, this is because the developed model includes warehouse capacity as a barrier.

Analisa Perencanaan Kebutuhan Material pada Perusahaan Manufaktur Kertas dengan Metode Material Requirement Planning (MRP)

The development of the business world is currently experiencing a very rapid growth rate. Planning for the proper procurement of materials will be able to facilitate the company's production activities. Production process activities in companies often experience problems such as inaccurate production, which is caused by delays in the arrival of raw materials, and damage to raw materials due to too long storing in warehouses. This study aims to plan future needs with the Material Requirement Planning (MRP) approach by doing forecasting, making the Master Production Schedule (MPS), Material Requirement Planning (MRP) and conducting a cost analysis of the five methods used. The results of this study are forecasting product demand will be used linear regression method, and effective material requirements planning method with the lowest cost is to use the Lot For Lot (LFL) method of $ 164.48. The application of the Lot For Lot method can reduce costs by 90.06% compared to the Fixed Period Requirements (FPR) method used by the company.

Towards forecast techniques for business analysts of large commercial data sets using matrix factorization methods

This research article suggests that there are significant benefits in exposing demand planners to forecasting methods using matrix completion techniques. This study aims to contribute to a better understanding of the field of forecasting with multivariate time series prediction by focusing on the dimension of large commercial data sets with hierarchies. This research highlights that there has neither been sufficient academic research in this sub-field nor dissemination among practitioners in the business sector. This study seeks to innovate by presenting a matrix completion method for short-term demand forecast of time series data on relevant commercial problems. Albeit computing intensive, this method outperforms the state of the art while remaining accessible to business users. The object of research is matrix completion for time series in a big data context within the industry. The subject of the research is forecasting product demand using techniques for multivariate hierarchical time series prediction that are both precise and accessible to non-technical business experts. Apart from a methodological innovation, this research seeks to introduce practitioners to novel methods for hierarchical multivariate time series prediction. The research outcome is of interest for organizations requiring precise forecasts yet lacking the appropriate human capital to develop them.

Information System Prediction With Weighted Moving Average (WMA) Method And Optimization Distribution Using Vehicles Routing Problem (VRP) Model for Batik Product

Distributing the products evenly to the customers is the problem faced by the producer of batik product. The purpose of this research is to optimize product distribution using demand prediction data from each store. The way of distributing products affects products brand and revenue in society. To cover this problem we propose the method to fulfil customers demand and to promote the brand by using Weighted Moving Average (WMA) to forecast product demand in each distributor store. To deliver the product for the customer, we are using the model of Vehicle Routing Problem (VRP) and Sequential Insertion Algorithm. We obtained the result that both distribution cost and travel time of distribution is successfully reduced. The results obtained that the WMA method can predict product demand by generate the small error value with MAPE the result is 0 until 17% which classified as very accurate-good. For distribution optimization using the VRP model and Sequential Insertion algorithms the result of optimal costs is IDR 36244, with a previous cost is IDR 150.000 in one month so can reduce 75%. Besides that, the travel time is 130.05 minutes with the distance 69.7 km, and routes that have been optimized.

Forecasting Product Demand in Terms of Demand Volatility

Forecasting Product Demand in Terms of Demand Volatility

Chinese residents’ cold chain logistics demand forecasting based on GM (1,1) model

Cold chain logistics is a complex system, involving the whole process of temperature control of the entire chain from the packaging, transporting, storage and consumption of the products. The meeting of the market demand must be based on the forecasting. The main goal of this study is to forecast product demand values for cold chain logistics from 2010 to 2015 in China using GM (1,1). The residents’ consumption data of refrigerated and frozen food ranging from meat to milk were obtained from the China statistical yearbook of 2005 to 2009. A gray forecast equation of cold chain logistics demand is established and estimated by DPS 7.05 software so as to confirm its scientific and accurate forecasting system. The results show all the prediction accuracy of GM (1,1) model of the refrigerated and frozen food demand of historical data reached level 1 except average residual error, which reached level 2. So, we can see that the previous established GM (1,1) models have achieved good prediction effect. Meanwhile, by 2015, the amounts of products transported under refrigeration condition will be more than 55,649 tons, with a 4.3 proportion growth per year in the period of 2010 to 2015. Key words: Cold chain logistics, demand forecasting, grey model, China.

Forecasting product demand of a bottling company in Ghana

Forecasting product demand is crucial to any supplier, manufacturer, or retailer; as the process determines the quantities that might be purchased, produced, supplied or shipped in future. For forecasting product demand of a Bottling Company in Kumasi (Ghana), seasonality indices were calculated to predict the demand levels of 3 products for the next 5 years based on the previous 5 years data. The results showed that one of the products enjoyed a higher volume of demand. The forecast data could be used to plan the production, in relation to peak and off-peak demand of the products; particularly, the period of the year for which production must be increased above normal. Journal of Applied Science and Technology Vol. 13 (1 & 2) 2008: pp. 122-128

Product Demand Forecasting with a Novel Fuzzy CMAC

Forecasting product demand has always been a crucial challenge for managers as they play an important role in making many business critical decisions such as production and inventory planning. These decisions are instrumental in meeting customer demand and ensuring the survival of the organization. This paper introduces a novel Fuzzy Cerebellar-Model-Articulation-Controller (FCMAC) with a Truth Value Restriction (TVR) inference scheme for time-series forecasting and investigates its performance in comparison to established techniques such as the Single Exponential Smoothing, Holt's Linear Trend, Holt-Winter's Additive methods, the Box-Jenkin's ARIMA model, radial basis function networks, and multi-layer perceptrons. Our experiments are conducted on the product demand data from the M3 Competition and the US Census Bureau. The results reveal that the FCMAC model yields lower errors for these data sets. The conditions under which the FCMAC model emerged significantly superior are discussed.

Heuristic Interpretation of Infeasible Solutions to a Production Scheduling Problem in Cigarette Manufacture

In this paper an end-user Infeasibility Resolution Heuristic (IRH) is developed to assist managers and production schedulers in the evaluation and interpretation of infeasible solutions arising from the use of a production scheduling model of a tobacco processing plant. The primary aim of the model is initially to determine whether the forecast annual demand for tobacco products can be met by the plant, and if not what must be done. Having determined that the demanded quantities are feasible, the model is then used to arrive at a production schedule such that the minimum number of machines are used in the entire process for the planning period (the model is a monthly one linked together by the closing stock equations for up to 12 months ahead). The operations of the plant modelled include the selection of the appropriate blend of leaf, assignment of the forecast product demand to suitable making machines, production of the required quantity of filters, assignment of the cigarettes produced to specific packers and then the determination as to whether the final product produced in the current period will be used to satisfy the current month's demand or demand in the future. The problem of sequencing the monthly production once the production schedule is determined has not been dealt with in this paper.

The indexes of consumer sentiment and confidence: Leading or misleading guides to future buyer behavior

The University of Michigan Survey Research Center's indexes of consumer sentiment and expectations and the Conference Board's index of consumer confidence and expectations are analyzed using time-series methods. The indexes were examined for a priori information content about a number of important variables associated with forecasting product demand and aggregate buyer behavior patterns. The empirical findings suggest that the indexes are useful in predicting future values of aggregate consumer spending and business and economic activity. The Michigan measure appears to outperform the Conference Board measure when predicting durable good purchases, whereas the reverse is true for general economic activity. Both measures are useful for predicting business activity.

Measuring the Quantity and Mix of Product Demand

A research approach is presented for situations in which a market planner must forecast product demand at alternative price levels, but where historical market data are irrelevant. An empirical study, in the telephone market, is presented to illustrate and demonstrate the effectiveness of the approach.