Case Study of Fuel Consumption by Vehicles Utilizing the Postulates of Bounded Rationality

Joseph E. Mullat * (PDF)

Abstract

The article presents a new method called "Blind Data Analysis", which simplifies the analysis of numerous statistical indicators with an unknown distribution. It highlights the practical context as well as the need for a rational categorization to improve the reliability of forecasts. It seems that this method makes it more accessible to consider indicators, following the principle of parsimony, similar to Ockham’s Razor. It explores and discusses the basic postulates of bounded rationality in decision-making process. Arguments based on postulates include a direct justification for the reliability of the method based on the additional postulate of a “constraint” of reality; this seems to be a more dynamic method. The study uses this method with data from the Spritmonitor.de database, which focuses on vehicle information, including gas and electricity consumption, and vehicle mileage. The database helps users track their fuel savings and related costs by providing thousands of car users with real-world cost per gallon or liter per 100 km. The postulates of rationality were tested against this database using the Excel macro program.

Keywords: data analysis; decision-making; vehicle; fuel consumption; monotonic system