<![CDATA[Air transportation is highly favored for its time efficiency and comfort, especially on busy routes such as Surabaya–Jakarta. However, the dynamic fluctuation of airline ticket prices often makes it difficult for consumers to plan their trips. This study aims to develop a predictive model for airline ticket prices on the Surabaya–Jakarta route using the multiple linear regression method. A total of 10,000 rows of data were analyzed using statistical approaches and analytical processes based on Google Colaboratory, involving stages such as data import, preprocessing, variable transformation, data splitting (training and testing), and classical assumption testing. The resulting regression model demonstrated excellent performance with an R-squared value of 96.4%, indicating that most of the price variation could be explained by independent variables such as airline, departure time, travel duration, baggage capacity, and service type. Violations of assumptions such as normality and heteroskedasticity were addressed through logarithmic transformation and the use of regression with robust standard errors. Furthermore, multicollinearity was minimized using Ridge Regression. Model evaluation showed no signs of overfitting and produced stable prediction results. Only a few variables were statistically significant, highlighting the importance of analyzing variable contributions to enhance model efficiency. The predictive model developed in this study provides accurate and practical results, making it useful for consumers in travel planning and for airlines in developing more competitive pricing strategies.]]>
