The dynamics of flight price predictions
-Flight pricing models have long been of interest to researchers, -airline companies, and consumers alike (Sun et al. 2024). As global air -travel expands, so do the complexities involved in predicting flight -prices due to the dynamic nature of factors such as fuel costs, demand -fluctuations, and the regulatory environment (Sun -et al. 2024; Belobaba, Odoni, and Barnhart 2015; Association et al. -2019; Borenstein and Rose 1994). Predicting flight prices -accurately allows both passengers and airlines to optimize travel -schedules, with potential economic and environmental benefits (Belobaba, Odoni, and -Barnhart 2015). In recent years, machine learning techniques have -been widely adopted in various industries, including aviation, to model -complex relationships and predict outcomes such as pricing (Sherly -Puspha Annabel et al. 2023; Kalampokas et al. 2023; Xu et al. -2021). For this project, we propose a machine learning-based -predictive model designed to estimate future flight prices based on -historical data from domestic flight routes to and from India.
-The importance of predicting flight prices
-Accurately predicting flight prices has significant implications for -multiple stakeholders. For consumers, anticipating price trends helps -optimize travel costs, allowing for better budgeting and planning (Association -et al. 2019; Borenstein and Rose 1994; Gillen and Morrison 2005). -Airlines, on the other hand, can improve their revenue management -strategies through dynamic pricing, ensuring that flights operate closer -to capacity while adjusting pricing to meet seasonal demand shifts and -market competition (Belobaba, -Odoni, and Barnhart 2015; Borenstein and Rose 1994; Gillen and Morrison -2005). Additionally, by integrating environmental factors such as -carbon emissions into the pricing model, this research can help airlines -design more sustainable routes, potentially leading to fewer emissions -through better traffic management and efficient flight operations (Sun -et al. 2024; Association et al. 2019; Gillen and Morrison 2005; Xiong et -al. 2023). The aviation industry contributes significantly to the -global economy but also poses challenges due to its substantial carbon -footprint (Brueckner and Abreu 2017).
-Methodological approach and dataset
-This project utilizes a Kaggle dataset on domestic flights in India, -covering variables such as airline name, total stops, departure and -destination locations, and flight prices [3–5, 8]. Using regression -techniques within a machine learning framework, the researchers aim to -identify which variables most significantly impact flight prices. -Initial findings suggest that factors such as the number of stops, -flight duration, and seasonal demand (correlating with major holidays in -India) are key drivers of price variation. For instance, a notable spike -in flight prices is observed around holidays in India, are insights that -not only advance knowledge in transportation economics but also serve as -a crucial decision-making tool for airline pricing strategies (Sun -et al. 2024; Association et al. 2019; Borenstein and Rose 1994; Gillen -and Morrison 2005).
-Multidisciplinary implications
-By developing this predictive model, the research bridges the gap -between economics, machine learning, and environmental sustainability. -Future applications of this work could extend beyond pricing into areas -like route optimization and emission reduction strategies (Association et -al. 2019). As flight prices are a key variable in both economic -and environmental calculations, understanding their underlying dynamics -can contribute to more sustainable and cost-effective air travel in the -future (Belobaba, -Odoni, and Barnhart 2015; Association et al. 2019; Gillen and Morrison -2005).
-