In response to evolving technologies, the use of business intelligence (BI) and big data has become essential for organizations to enhance decision-making and achieve objectives. It also helps companies to anticipate trends and allocate resources efficiently. This project discovered several challenges encountered by a service-based company in the F&B industry, including limited visibility into sales processes, difficulties in forecasting customer demand, and inefficiencies in estimating inventory during operations management. Thus, this project aimed to optimize the company’s operations by applying predictive analytics (PA), classification techniques, and cross-validation methods within the CRISP-DM methodology, using sales data to develop models for predicting sales, inventory, and customer demand. To achieve this, three algorithms which are Decision Tree (DT), Naïve Bayes (NB) and Random Forest (RF) were chosen for their robustness and simplicity. After experimentation, Random Forest emerged as the best-performing algorithm, achieving an accuracy of 54.06% for sales prediction, 88.54% for inventory prediction, and 95.47% for customer demand prediction. A comprehensive dashboard was created to visualize these results, improving decision-making, reducing waste, and enhancing efficiency. The dashboard was validated by three experts, who evaluated its usefulness, ease of use, ease of learning, and satisfaction. All experts agreed the dashboard was well-executed, emphasizing its user-friendliness and clarity and can be quickly mastered. The integration of these models into service operations has proven valuable in driving business growth and efficiency.