There are a lot of hazardous object movements on PTTEP sites, especially vehicles and machineries. The environment inherently leads to more likelihood of accidents or near-misses if poor safety management is in place. The company decided on using technology to identify dangerous objects and humans to report dangerous cases when humans enter a danger object's hazardous areas. Forklift detection occurs to detect forklifts in working areas, detect humans in prohibited areas, and report any positional violation. Customers can use this data for safety training to increase safety awareness and improve working environment safety.

CCTVs are installed in the working area; footage from the CCTVs can be used to train and create the detection model. Solution steps - Build the detection model: Train the detection model by obtaining an image from CCTV and marking it with a forklift to use the train detection model. - Perspective banned area: after obtaining the forklift's bounding box, create a banned area for the forklift and display it in perspective view. - Integrate with other models: combine the forklift detection model with other detection models. - Database Connection - connect the forklift output to a database and display the results on a local website.

After installing forklift detection at the site, it is possible to detect people who are in close range to the forklift. The safety department or engineer in charge of the working environment can apply this data to instruct employees and use the application to track forklift incidents on a monthly schedule.