A Novel K-means-LSTM Hybrid Model of Supply Chain Inventory Management
Open Access
- 26 April 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 419-425
Abstract
With the rapid development of the internet and e-commerce, demand forecasting for products and inventory optimization has become especially critical. This study develops a hybrid model that combines the K-means clustering algorithm and Long Short-Term Memory (LSTM) networks, aiming to group the three-dimensional historical data of merchants, warehouses, and product demand volumes, thereby categorizing time series and improving the accuracy of demand forecasting. This paper utilizes the shipping volume of 1996 "merchant + warehouse + product" combinations provided by JD Group from November 1 to November 11, 2022, to predict promotional trends from June 1 to June 20, 2023. Firstly, the JD warehouse products are divided into ten categories according to the K-means algorithm. Then, an LSTM model comprising an input layer, an output layer, and two hidden layers is constructed to predict the demand volume for each product category, which is subsequently verified through R² assessment. Results show that the R² maintains around 90%, indicating that the prediction model is not only reliable but also effective. The in-depth analysis and precise forecasting in this study demonstrate the practical application potential of K-means clustering and LSTM networks in enhancing forecasting accuracy, proving the significant value and robustness of the model established in the data-driven demand prediction domain.Keywords
This publication has 14 references indexed in Scilit:
- Development of a Time Series E-Commerce Sales Prediction Method for Short-Shelf-Life Products Using GRU-LightGBMApplied Sciences, 2024
- A Practical End-to-End Inventory Management Model with Deep LearningManagement Science, 2023
- A hybrid deep learning framework with CNN and Bi-directional LSTM for store item demand forecastingComputers and Electrical Engineering, 2022
- K-Means Clustering Approach for Intelligent Customer Segmentation Using Customer Purchase Behavior DataSustainability, 2022
- China’s commercial bank stock price prediction using a novel K-means-LSTM hybrid approachExpert Systems with Applications, 2022
- Cluster-based demand forecasting using Bayesian model averaging: An ensemble learning approachDecision Analytics Journal, 2022
- Predictive and prescriptive performance of bike-sharing demand forecasts for inventory managementTransportation Research Part C: Emerging Technologies, 2022
- On the Application of ARIMA and LSTM to Predict Order Demand Based on Short Lead Time and On-Time Delivery RequirementsProcesses, 2021
- Time Series Risk Prediction Based on LSTM and a Variant DTW Algorithm: Application of Bed Inventory Overturn Prevention in a Pant-Leg CFB BoilerIEEE Access, 2020
- Machine learning for multi-criteria inventory classification applied to intermittent demandProduction Planning & Control, 2018