Currently, the application of various information system technologies is undoubtedly widespread across almost all companies. PT. PQRS, one of Indonesia's tire manufacturers, has implemented an order management information system technology that comprehensively manages the sales process and customer orders. However, it still falls short of fully enhancing the company's operational effectiveness. At times, there remains an excess in production for certain products due to sales results significantly below the set targets. This leads to stockpiling in warehouses, increased production costs, shrinkage, or expiration, and a decrease in selling prices, all of which contribute to significant company losses. The aim of this research is to predict the number of tire sales for specific products in the upcoming period using sales transaction data from an existing order management information system. Predictions are made by analyzing the accuracy of neural network algorithms, including the Gated Recurrent Unit (GRU), Multilayer Perceptron (MLP), and Long Short-term Memory (LSTM). In this study, sales predictions for the MT000275 product were based on sales data from January 1, 2003, to December 31, 2022, utilizing the 5-Split Cross-Validation test. Notably, there were significant anomalies in sales fluctuations from August 2020 to December 2022. The Long Short-term Memory (LSTM) algorithm demonstrated the most consistent and superior performance, exhibiting fairly good accuracy with an average accuracy value of 94.36%, or equivalently, an average error value of 5.64%. These tire sales predictions can provide a more accurate basis for determining optimal tire production quantities, thus helping to prevent company losses. The findings of this research provide valuable insights for manufacturing companies in improving production planning, minimizing inventory overstock, and supporting data-driven decision-making processes. Furthermore, the proposed forecasting approach can contribute to the development of intelligent sales prediction systems in industrial supply chain management