Paper Contents
Abstract
Street vendors, such as food stalls, fruit sellers, and garment vendors, often experience fluctuating daily incomes based on natural factors such as sunlight and time of day. This study explores how environmental variables, particularly shadow length, sunlight hours, and time of day, influence vendor earnings. The shadow length was calculated using the solar elevation angle, which reflects the availability of direct sunlight at the location of the stall. Sunlight hours were measured as the cumulative exposure during working hours, whereas the time of day was encoded hourly to capture customer flow patterns. Daily income is the target variable, with weekday versus weekend classification serving as a control to account for differences in consumer behavior. The methodology involves collecting observational data from vendors over several weeks and recording their income, alongside environmental parameters derived from solar geometry and temporal factors. A machine learning model, specifically a multilayer perceptron (MLP), was trained to identify the nonlinear relationships between sunlight exposure patterns and vendor income. By humanizing these insights, this study highlights how a simple aspect, such as the angle of the sun, can directly influence sales, customer presence, and vendor livelihoods. The findings are expected to support local vendors in optimizing their working hours and stall positioning to improve their earnings.
Copyright
Copyright © 2025 Mari Shankar S. This is an open access article distributed under the Creative Commons Attribution License.
