Application of an Advanced Dijkstra Algorithm for Automation Systems of 3D Piping Routing

As engineering, procurement, and construction (EPC) plant projects increase in size and complexity, the importance of accurate design in a short time frame is rising. In particular, the automation of piping route design is essential to shorten the project schedule and reduce the cost. This study aims to develop an automated pipeline route design model and proposes an advanced Dijkstra algorithm. First, digital image processing technology was applied to preprocess the data. In addition, 3D model data of plant equipment was set up to realize the developed model in 3D. The Advanced Dijkstra algorithm proposed in this study for piping design automation consists of three algorithms. First, the Dijkstra algorithm was applied to find the shortest route, and the Routing Priority algorithm, which considers the importance of design, was added to optimize the route. In addition, the Interference Obstacle Avoidance algorithm was used to consider errors caused by interference between equipment and pipes. The automatic piping route design model developed in this study was verified using the project data of the 'D' petrochemical plant in Korea. As a result of comparing the Bill of Material (BoM) extracted through the developed model with the BoM of the actual completed project, the matching accuracy was 98.9% in weight and 84.9% in length for the pipe quantity. The automatic pipe route design model is expected to store project data to manage advanced design know-how and contribute to sustainable growth in the field of pipe design.