Product Demand Forecasting

Project Code :TCMAPY1087

Objective

The primary objective is to rigorously evaluate the effectiveness of ARIMA, SARIMA, and LSTM models in predicting product demand. By employing diverse datasets and considering multiple influencing factors, the project aims to provide actionable insights for practitioners and decision-makers. The goal is to aid in the selection of the most suitable and accurate model, facilitating informed decision-making processes and optimizing resource allocation strategies.

Abstract

Forecasting product demand plays a pivotal role in optimizing inventory management and meeting customer needs. This study explores and compares three distinct time series models for demand forecasting: ARIMA, SARIMA, and LSTM. Autoregressive Integrated Moving Average (ARIMA) models the temporal dependencies in the data, while Seasonal ARIMA (SARIMA) incorporates seasonal trends for improved accuracy. Long Short-Term Memory (LSTM), a type of recurrent neural network, captures complex sequential patterns. Through rigorous analysis and evaluation, this research investigates the effectiveness of these models in predicting product demand. Real-world datasets are utilized to assess the predictive capabilities, considering various factors such as seasonality, trend, and irregularities in demand patterns. Insights derived from this comparative analysis aim to guide practitioners and decision-makers in selecting the most suitable model for accurate and efficient product demand forecasting, thereby facilitating informed decision-making and resource allocation.

Keywords: ARIMA, SARIMA, LSTM

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.