Inventory Forecasting Research Articles

Convolutional Neural Network and Bidirectional Long Short-Term Memory (CNN-BiLSTM)-Attention-Based Prediction of the Amount of Silica Powder Moving in and out of a Warehouse

Raw material inventory control is indispensable for ensuring the cost reduction and efficiency of enterprises. Silica powder is an essential raw material for new energy enterprises. The inventory control of silicon powder is of great concern to enterprises, but due to the complexity of the market environment and the inadequacy of information technology, inventory control of silica powder has been ineffective. One of the most significant reasons for this is that existing methods encounter difficulty in effectively extracting the local and long-term characteristics of the data, which leads to significant errors in forecasting and poor accuracy. This study focuses on improving the accuracy of corporate inventory forecasting. We propose an improved CNN-BiLSTM-attention prediction model that uses convolutional neural networks (CNNs) to extract the local features from a dataset. The attention mechanism (attention) uses the point multiplication method to weigh the acquired features and the bidirectional long short-term memory (BiLSTM) network to acquire the long-term features of the dataset. The final output of the model is the predicted value of silica powder and the evaluation metrics. The proposed model is compared with five other models: CNN, LSTM, CNN-LSTM, CNN-BiLSTM, and CNN-LSTM-attention. The experiments show that the improved CNN-BiLSTM-attention prediction model can predict inbound and outbound silica powder very well. The accuracy of the prediction of the inbound test set is higher than that of the other five models by 7.429%, 11.813%, 15.365%, 10.331%, and 5.821%, respectively. The accuracy of the outbound storage prediction is higher than that of the other five models by 14.535%, 15.135%, 1.603%, 7.584%, and 18.784%, respectively.

Inventory Prediction Using a Modified Multi-Dimensional Collaborative Wrapped Bi-Directional Long Short-Term Memory Model

Inventory prediction is concerned with the forecasting of future demand for products in order to optimize inventory levels and supply chain management. The challenges include demand volatility, data quality, multi-dimensional interactions, lead time variability, seasonal trends, and dynamic pricing. Nevertheless, these models suffer from numerous shortcomings, and in this research, we propose a new model, MMCW-BiLSTM (modified multi-dimensional collaboratively wrapped BiLSTM), for inventory prediction. The MMCW-BiLSTM model reflects a considerable leap in inventory forecasting by combining a number of components in order to consider intricate temporal dependencies and incorporate feature interactions. The MMCW-BiLSTM makes use of BiLSTM layers, collaborative attention mechanisms, and a multi-dimensional attention approach to learn from augmented datasets consisting of the original features and the extracted time series data. Moreover, adding a Taylor series transformation allows for a more precise description of the features in the model, thus improving the prediction precision. The results show that the models make the least mistakes when they use the AV demand forecasting dataset, with MAE values of 1.75, MAPE values of 2.89, MSE values of 6.76, and RMSE values of 2.6. Similarly, when utilizing the product demand dataset, the model also achieves the lowest error values for these metrics at 1.97, 3.91, 8.76, and 2.96. Likewise, when utilizing the dairy goods sales dataset, the model also achieves the lowest error values for these metrics at 2.54, 3.69, 10.39, and 3.22.

Analisis Peramalan dalam Menentukan Perencanaan Persediaan pada Toko Seragam Sekolah Hermul Jaya

Forecasting is an estimation of future events based on systematically compiled past data using certain methods. Forecasting is necessary as a basis for planning, therefore an appropriate forecasting method is needed to see a lower level of error so that inventory planning can be optimal and efficient. This study aims to find a more suitable method to be used at Hermul Jaya Clothing Store to determine the predicted amount of stock in the store. This research uses a descriptive quantitative approach and POM QM application to calculate the inventory forecast of uniform stores based on two indicators, namely Moving Average, Weighted Moving Average and Expponential Smoothing. From the results of this study, the stock inventory at Hermul Jaya Uniform Store is not yet effective because stock purchases or inventory are made without prior planning or it can be said that this store does not use forecasting methods

Analisis Peramalan Persediaan Bahan Baku dan Penentuan Stok Penyangga (Buffer Stock) Udang Vaname (Littopenaeus Vannamei)

This study examines the analysis of raw material inventory forecasting and buffer stock of vaname shrimp at one of the fishery companies in Makassar Industrial Estate (PT Bogatama Marinusa Makassar). The analysis method used to determine the supply of raw materials needed by the company is the ARIMA Box-Jenkins method. This method is used to forecast raw material inventory on time series data. Determination of buffer stock is done using standard deviation and policy factors. Vaname shrimp (Littopenaeus vannamei) raw material data was obtained from 2020 to 2022 (156 weeks). The results showed that the highest amount of raw material inventory occurred in October week 147 in 2022, amounting to 152,792 tons, while the lowest amount of raw material inventory occurred in May week 122 in 2022, amounting to 13,102 tons. The best vaname shrimp raw material inventory model is the ARIMA (1,1,1) model which has a Sum Square Error (SSE) value of 1.70250, Mean Square Error (MSE) value of 0.0112007. This model is used to forecast raw material inventory for the next 48 weeks. The forecasting results show that there will be a decrease in week 1 to week 6 and a relative increase in raw materials in week 7 to week 48 with a MAPE value of 4.54%. The amount of buffer stock that must be owned by the company is 37.311 tons.

Inventory Forecasting for Medical Environmental Protection Equipment Enterprises Based on BP Neural Network Optimized by Sparrow Search Algorithm

Deficiencies in corporate inventory forecasting are common problems, an inventory forecasting model using a sparrow search algorithm to de-optimize a BP neural network has been developed. Through the actual production of HX medical environmental protection equipment enterprises, the inventory of raw materials is studied, and the five factors that have the greatest impact on the inventory are selected. To establish an inventory forecasting model using the sparrow search algorithm to optimize the weights and thresholds of the BP neural network, and at the same time to compare and analyze the optimized model with the model before optimization. After comparing the results, it can be seen that the BP neural network optimized by the sparrow search algorithm is more stable and the prediction accuracy is more accurate than before, when predicting the inventory of the enterprise due to the optimization of the weights and thresholds.

Non-stationary time series data for natural rubber inventory forecasting: A case study

PPLK Corp. is a company that uses natural rubber as the main raw material to produce crumb rubber. The problem identified in PPLK Corp. is the insufficient amount of natural rubber received to produce and fulfill consumer demand. There have been fluctuations in the amount of natural rubber received and high variability between periods. To minimize this variability, it is necessary to forecast natural rubber requirements. The purpose of this study is to forecast the natural rubber inventory for the next periods using the best-fitted model, which is the Autoregressive Integrated Moving Average (ARIMA) method. A total of 547 daily data points from 2021 to 2022 were used. As a result, the ARIMA (1,1,2) model was found to be the best model for natural rubber forecasting in the rubber factory. The ARIMA (1,1,2) model had the smallest AIC value compared to others. The total daily natural rubber need is forecasted to be around 67.588 kilograms with a range between 64.805 and 70.421 kilograms per day. However, it should be noted that this study was limited to short-term forecasting only.

Inventory Forecasting Analysis using The Weighted Moving Average Method in Go Public Trading Companies

This research aims to analyze inventory forecasting using the weighted moving average method and then compare the trading companies' patterns. The research method used is quantitative descriptive with secondary data of inventory in the period 2018-2022 which provide quarterly. This research uses the weighted moving average method to calculate forecasting of inventory by Microsoft Excel data analysis techniques. This research shows the highest inventory forecasting on PT Sumber Alfaria Trijaya Tbk (AMRT) occurs in the first quarter of 2023 with the amount of 10.537.541 and the lowest forecasting occurs in the second quarter in 2023 with the amount of 10.431.677. The highest inventory forecasting on PT Erajaya Swasembada Tbk (ERAA) occurs in the second quarter of 2023 with the amount of 6.443.525 and the lowest forecasting in the fourth quarter of 2023 with the amount of 6.418.659. The highest inventory forecasting on PT United Tractors Tbk (UNTR) occurs in the third quarter of 2023 with the amount of 12.239.422 and the lowest forecasting in the first quarter of 2023 with the amount of 12.050.681. Based on the study's results, the tracking signal value at AMRT was 2,17, ERAA was 0.01, and UNTR was -0.08. The three companies' results prove that the weighted moving average can be used to determine inventory forecasting for the next period because the tracking signal value is still within the control limits of ±4.

Analysis of E-Commerce Product Sales Prediction Based on Ga-Bp Neural Network

In recent years, the market for the sale of various goods has been expanding and with it comes huge amounts of data that needs to be specifically processed and analyzed. For a company, the adoption of artificial intelligence algorithms with better performance in handling big data is an important step in influencing the future decisions and future growth of the company. In this paper, an optimized neural network based on GA-BP is constructed to analyze samples with different characteristics to predict the future sales in the short term. After testing, the prediction results are found to be good. First, the degree of influence between the indicators is examined using BP neural network. By comparing whether the inventory indicator at the beginning of the month is used as the model input, the error of the end-of-month inventory forecast content is obtained, which leads to the conclusion that the inventory at the beginning of the month has a greater impact on the inventory at the end of the month. Similarly, it can be concluded that the commodity type has a greater impact on the commodity pool. Second, to further analyze the relationship between commodity type and commodity inventory, the Eta test was used. It was found that item type A2 had the greatest impact on merchandise inventory, followed by type A1 and finally type A3. This finding is consistent with the characteristics of the dataset. Considering that the optimized GA-BP neural network model is mainly used to predict merchandise sales in the next 3 months; secondly, considering the specificity of some data in the sample, it is not suitable to use the neural network for analysis and prediction. For all data, the data that could be analyzed by GA-BP were found by gradient descent method, and it was found that the data with target value below 5500 had the best prediction effect. Finally, according to the given accuracy evaluation index, the score of the model is calculated as 0.678 5, and the model works well.

Inventory Forecasting and Control Decisions for Effective Inventory Management in the South African Automotive Component Manufacturing Industry: Pre COVID-19 and Lockdown Period

The COVID-19 global health pandemic significantly affected the global economy and the automotive industry when lockdown measures were implemented. This study seeks to investigate the influence of inventory forecasting and control decisions on effective inventory management in the South African automotive component manufacturing (SAACM) industry. Using a positivistic paradigm with a quantitative research approach, data was sourced from 162 Automotive Component Manufacturers (ACMs) to establish whether their inventory forecasting and control decisions changed from prior COVID-19 to the lockdown period during the COVID-19 pandemic. The multiple regression analysis found statistically significant relationships between inventory forecasting decisions and effective inventory management prior to COVID-19, inventory control decisions and effective inventory management prior to COVID-19 and inventory forecasting decisions and effective inventory management during the lockdown period. Thus, regardless of the COVID-19 pandemic, inventory managers in ACMs in South Africa (SA) should use inventory forecasting methods such as demand forecasting, economic order quantity and materials requirement planning. They should further consider using an inventory information sharing system and inventory replenishment procedure to manage inventory more effectively.

Primena kvantitativnih metoda prognoziranja zaliha u lancima snabdevanja na primeru kompanije "Leoni"

In order to effectively manage the supply chain, it is important to effectively manage the inventory within it. The application of quantitative forecasting methods contributes to the accuracy and validity of forecasting and defining stock flows. The subject of this paper is to review the use of quantitative methods of inventory forecasting, namely exponential smoothing, adaptive filtering, the method of moving averages and regression analysis. The aim of the paper is to show inventory forecasting using the mentioned quantitative methods on the example of the company "Leoni Wiring System Prokuplje". The results of the research indicate that applied quantitative methods of forecasting in the supply chain provide reliable and accurate forecasts on the basis of which material orders can be defined in order to ensure the smooth functioning of production, on the example of the company "Leoni Wiring System Prokuplje".

Application of The Combined Method in Inventory Forecasting Electricity at PT PLN (Persero) ULP Sibuhuan

The forecast of electricity consumption is a prediction of the use of electricity for the future by referring to the use of electricity in past data. Estimates of electricity consumption needs are intended to estimate how much electricity consumption is used by customers and must be provided by PT PLN (PERSERO) ULP SIBUHUAN as a provider of electrical energy services. Population growth occurs along with the development of an area, so it affects the demand for electricity and the need for electricity consumption. Load mapping must be done to maintain the continuity and distribution of electrical energy to customers. One way to preserve the continuity of service is to estimate electric power consumption using several forecasting methods. One of the methods that can be used is to combine several load forecasting methods called the combined method. The combined method is a model that incorporates various estimation methods, including econometrics, analytics, and trends, using historical customer data originating from BPS (Central Statistics Agency) Padang Lawas Regency and PT. PLN (PERSERO) ULP SIBUHUAN. The estimated load between 2021 and 2025 is 625,070,452 kWh, with an increase in electricity consumption of 20.7%, household expenses increasing by 1.8%, and business sector expenses by 1.35%. The method used is to calculate manually using Microsoft Excel to obtain forecasting values for population growth; increasing electricity consumption is very important for planning electric power generation, development, and planning of electric power distribution and mapping of loads on customers.

Comparative analysis of ARIMA and double exponential smoothing for forecasting rice sales in fair price shop

One of the most challenging issues during the pandemic is managing uncertainties in demand, customer behavior, and market trends. Such instability and unpredictability resulted in numerous cases of excess stock when demand declined or a shortage of commodities when demand for certain goods increased significantly. The research presented in this paper contributes to modelling and forecasting rice sales demand in a Fair Price Shop (FPS) in Kerala, India by employing a time series technique. Our research shows how past demand data can be used to estimate future demand and how these forecasts impact the Public Distribution System (PDS). Our study employs Autoregressive Integrated Moving Average (ARIMA) and Double Exponential Smoothing (DES) techniques to develop future prediction models that significantly increase the efficiency and accuracy of demand and inventory forecasting. The forecast models generated from past data are verified and validated in the real case application using the Mean Absolute Percentage Error (MAPE) that helps to forecast the demand of inventory required in FPS. The proposed ARIMA and DES outperform the forecasts made by the empirical model, with ARIMA doing better in terms of future forecasts.

Implementation of supply management strategies by the pharmacy service in a general hospital during the COVID-19 pandemic

ObjectiveGiven the global uncertainty faced due to the Covid-19 pandemic, health services were forced to adjust inventory management and purchase projections. This publication aims to describe the strategies taken and their impact on the supply chain indicators by the pharmacy service in the management of drug purchases during the pandemic to expose the importance of pharmacist in charge of the supply chain. MethodsThis observational study describes the drug purchasing system in a general hospital and the strategies used to manage drug supply. The actions proposed by the pharmacy department are listed chronologically related to inventory issues and purchasing decisions. The accuracy of the purchase forecast was evaluated by calculating indicators such as the mean absolute standard deviation (MAD) and the variance of the forecast error (MSE). Inventory days and inventory turnover indicators were also compared in pre-pandemic and pre-pandemic periods. FindingsIn general, the forecast error given by MAD and MSE tended to decrease. Specifically, from the 82 drug categories, during the pandemic period, this indicator decreased in 72 (88%), increased in 9 (11%), and remained the same in only 1 (1%) of the categories. In financial terms, comparing the 2018–2019 and 2020–2021 periods, a favorable result was obtained when evaluating the inventory turnover indicators, which decreased by 0.01 points and the days of inventory increased on average by two days. ConclusionsThe implementation and use of these indicators prevented drug shortages, reducing inventory forecast errors. A pharmacist with knowledge in inventory management allows managing a process of continuous improvement and tactics for efficient inventory management without neglecting the benefit to patients or the economic profitability of the service. There were limitations since digital operating systems do not generate centralized or organized data for this type of analysis.

Rethinking Inventory Forecasting Problems in E-commerce: Exploring the effect of integrating forecasting and inventory decisions

Forecasting for inventory control is the process of calculating the inventory needs to fulfill future consumer demand. In general, this process is divided into two sub-processes. The first sub-process receives the current inventory information and forecasts future information, e.g. forecasts future demand from the demand information in the past. The second sub-process uses the forecast information as input to make inventory decisions, e.g. use a product demand forecast to decide how many units of this product to buy. Recent works highlight the importance of integrating forecasting with final inventory decisions, however, there is very little empirical evidence to support that integrating the decision is the best solution. In this work, we propose to explore the effect of integrating the inventory decision into the forecasting problem and compare it with the state-of-the-art approaches. For this, we evaluated the approaches in different operational tasks belonging to our business. Our preliminary findings show that predicting operative decisions instead of demand information could be better and the benefit can be capitalized even in low data scenarios.

Substation Equipment Spare Parts’ Inventory Prediction Model Based on Remaining Useful Life

A large variety of high-value substation relay protection equipment occupies a considerable amount of inventory space and capital in electric power companies. To improve this problem, this study proposes an inventory prediction model based on the remaining useful life (RUL) of equipment. The model acquires the RUL data of equipment by using the support vector regression (SVR) algorithm, and then, by taking this data as the main factor and the environmental factors and human factors during the operation of equipment as secondary factors, the model can realize the prediction of relay protection equipment in the substation. At the same time, the nature of the enterprise and the requirements for safety inventory are considered. The comparison of calculation results and error analysis, as well as the calculation time, all indicate that the RUL-based inventory forecasting is the best one. This model not only has high prediction accuracy but also has strong stability and portability. The model can provide a strong decision basis for improving the inventory management of the enterprise, enhancing the resource allocation capability, and formulating the spare parts procurement plan under the condition that the spare parts inventory reaches the safety stock.

The Application of Neural Networks in the Construction of a Program for the Markup of Text

This paper explores the use and applications of neural networks in the construction of a text markup program. This research paper describes various tasks that require textual data analysis and discusses the problems, issues, and solutions that accompany them. Interest in neural networks has increased in recent years, and they are find­ing applications in a wide variety of fields, such as business, medicine, engineering, geology, and physics. Neural networks have made great strides in forecasting, planning, and management. There are several reasons for this situation. Neural networks are very powerful modeling systems capable of creating complex dependencies. Neural networks can be widely used in areas such as text/speech recognition, semantic search, decision support/expert systems, inventory forecasting, data storage systems and content analysis. The object of the work is the process of functioning of neural networks and an algorithm for text markup recognition. The purpose of the scientific paper is the application of neural networks in the construction of the program for the markup of the text. In order to achieve the goal, the following tasks were put forward: a) study existing neural networks, b) choose a neural network to cre­ate a model and study its structure, c) convert input data to feed it into a neural network model. Representation and analysis of symbolic structures in neural networks seems to be an interesting and useful direction in neural network theory. In this paper, we have reviewed some neural network architectures that may merit further consideration in these circumstances. In what follows, we will focus on specific experiments in this area. Thus, this paper outlines a problem area for further research and testing.

Data-Driven Inventory Forecasting in Base Stock Systems Calibrated with a Fill Rate Constraint

Data-Driven Inventory Forecasting in Base Stock Systems Calibrated with a Fill Rate Constraint

Application of Weight Moving Average Method in Forecasting Drug Inventory Application at Dermayu Health Center

The drug inventory forecasting application at Dermayu Public Health Center is an application that is used to predict the amount of drug inventory for the next month based on an analysis conducted on the data on the amount of drug inventory for the last 12 months. In addition, in this application there are 2 (two) methods used for comparative analysis of drug inventory forecasting, namely the Weight Moving Average Method and the Linear Regression Method. Where the comparison of the two methods is seen from the forecasting results and the level of accuracy of the forecast results. With the forecasting results in these 2 (two) methods, it can be used as a reference for the health center in making decisions to manage the amount of drug inventory each month. The drug supply forecasting application was created using the Visual Basic .Net programming language with SQL Server database. Based on the tests that have been carried out, it can be concluded that the functional of the application for forecasting drug supplies at Dermayu Health Center has been running as it should, and the drug inventory forecasting application is able to display the results of forecasting the amount of drug inventory for the next month through 2 methods, namely the Weight Moving Method. Average and Linear Regression Method..

On the Application of ARIMA and LSTM to Predict Order Demand Based on Short Lead Time and On-Time Delivery Requirements

Suppliers are adjusting from the order-to-order manufacturing production mode toward demand forecasting. In the meantime, customers have increased demand uncertainty due to their own considerations, such as end-product demand frustration, which leads to suppliers’ inaccurate demand forecasting and inventory wastes. Our research applies ARIMA and LSTM techniques to establish rolling forecast models, which greatly improve accuracy and efficiency of demand and inventory forecasting. The forecast models, developed through historical data, are evaluated and verified by the root mean squares and average absolute error percentages in the actual case application, i.e., the orders of IC trays for semiconductor production plants. The proposed ARIMA and LSTM are superior to the manufacturer’s empirical model prediction results, with LSTM exhibiting enhanced performance in terms of short-term forecasting. The inventory continued to decline significantly after two months of model implementation and application.

Development of a hybrid framework for inventory leanness in Technical Services Organizations.

Inherent uncertainties in demand and supply make it problematic for supply chains to accomplish optimum inventory replenishment, resulting in loss of sales or keeping excessive inventories. To cope with erratic demands, organizations have to maintain excessive inventory levels, sometimes taking up to one-third of an organization’s annual budget. The two most pressing concerns to handle in inventory management are: how much to order and when to order. Therefore, an organization ought to make the correct and timely decisions based on precise demand information to avoid excessive inventory accumulation resulting in enhanced competitive advantage. Owing to the significance of inventory control and analysis, this paper reports on developing and successfully implementing a hybrid framework for optimum level inventory forecasting in Technical Services Organizations. The proposed framework is based on a case study of one of Pakistan’s leading Technical Services Organization. The paper presents a statistical analysis of historical data and a comprehensive fault trend analysis. Both these analyses set a solid foundation for the formulation of a comparative analysis matrix based upon price and quantity based analysis of inventory. Finally, a decision criterion (Forecasting Model) is proposed using three primary forecasting techniques with minimum error calculations. The study’s finding shows a forecast error of 142.5 million rupees in the last five years, resulting in the accumulation of more than 25 thousand excessive inventory stock. Application of price and quantity based analysis identifies that 65% of the annual budget is significantly dependent upon only 9% (in terms of quantity) of "High Price and Small Quantity" Items (HS). These HS items are forecasted through three different forecasting methods, i.e., Weighted Moving Average, Exponential Smoothing, and Trend Projection, with Minimum Absolute Deviation to significantly reduce the forecasting error while predicting the future required quantity. The research work aims to contribute to the inventory management literature in three ways. First, a new comparative analysis matrix concept for identifying the most critical items is introduced. Second, a Multi-Criteria Forecasting Model is developed to capture a wide range of operations. Third, the paper suggests how these forecasting criteria can be integrated into a single interactive DSS to maintain optimum inventory level stock. Even though the DSS framework is based on data from a single organization, the application is expected to manage inventory stock in a wide range of manufacturing and services industries. This study’s proposed hybrid framework is the first of its kind that encapsulates all four dimensions of inventory classification criteria, forming a multi-criteria hybrid model within a DSS framework.