Computation Time Analysis of D16 Algorithm for Surface Water Flow Direction Using Decision Tree

Abstract

Digital Elevation Model (DEM) is data used for modeling water flow. DEM data is in the form of a grid with X and Y points as coordinates and Z representing the surface elevation. One of the challenges in modeling the direction of water flow is the computational time of the algorithm to model the flow on DEM data. Based on the initial tests conducted in this study, it can be seen that the increase in computation time will always be directly proportional to the amount of data processed by the application. To overcome this problem, certain methods are needed to shorten computation time to make it more efficient. This study contributes to the existence of a computational model using a decision tree to solve the problem of finding the direction of surface water flow using the D16 algorithm. The D16 algorithm is a method for modeling the direction of surface water flow. The algorithm works by forming 16 flow directions in the search space. For testing in this study, we will compare the sequential computational model and the computational model with the decision tree that the author made to see the computational time efficiency of the D16 algorithm completion model. The test results show that modeling the direction of water flow with a decision tree produces a good computation time compared to the sequential program model, with an average computational difference of 82.66 seconds.