Implementation of ANN in Predicting Construction Costs Considering Macroeconomic Factors

Budget calculations for residential construction is one of the crucial steps in development planning. Estimating the budget plan is not an easy task and requires a considerable amount of time to accomplish. Given the continuous growth in construction in Indonesia driven by macroeconomics, many projects necessitate cost estimation planning to determine the initial contract value of the construction project. The initial value of the construction project can be predicted using artificial intelligence methods. Producing an accurate predictive model is one of the goals to be achieved. The prediction methods employed in this study are artificial neural network (ANN) and linear regression (LR). Both methods will be evaluated to determine which one is best for predicting construction costs/m2. From the data collection, 119 pieces of data were obtained and utilized. Among them, 119 were projects spanning from 2012 to 2023. The results indicate that in the tested dataset, the ANN method yielded a MAPE value of 15% and an R value of 0.68. Based on the testing results, ANN emerges as the best method for predicting the construction cost/m2 of construction projects.