In the oil and gas industry, various machineries and equipment are used to perform oil and gas extractions. The problem arises when there is unplanned maintenance on any equipment. Unplanned maintenance will result in unplanned deferments that disrupt business operations. Companies may have developed monitoring systems based on current and historical equipment statuses, but ideally there should be mechanisms to conduct or produce real-time forecasts on equipment conditions. In this paper, linear regression models were tested and deployed in a system developed to forecast flow rate of seawater lift pumps of an offshore platform. Apart from identifying and evaluating a suitable statistical model to derive the forecasts, this paper presents a tool that was developed using the selected model to automate real-time data extraction and execute the prediction process. The models were developed based on raw data that were accumulated from an oil and gas company over a period of 3 months. Of the 3 months’ data, the first 2 months of data were used as the training data, and the last one month was used for testing the models. Data cleansing was performed on the dataset whereby unwanted values that could affect accuracy of the model or any other data with values not processable by the models were eliminated. Results indicated that Autoregressive (AR) model is suitable for a real-time prediction of an offshore equipment.