This study compares the Logistic and Gompertz models for fitting “S” curves in road infrastructure projects in Pichincha, Ecuador, aiming to identify the most accurate and efficient tool for predicting project progress and managing resources. Using Python and the SciPy library with Levenberg-Marquardt optimization, the models were applied to the construction data from 2008–2018 and validated with projects from 2018–2023. Results indicate that the Gompertz model outperforms the Logistic model in precision, with an R² increase from 0.7403 to 0.7649, compared to the Logistic model's 0.6957. The Gompertz model better captures early-stage growth and anticipates cost and progress dynamics, improving planning and reducing delays. Thus, the study recommends adopting the Gompertz model as the standard for project management in the region. Nonetheless, it underscores the need to complement mathematical forecasting with contextual factors such as market trends, resource availability, and project scope to ensure holistic and effective decision-making.
Hidalgo et al. (Sun,) studied this question.