International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 13 Issue: 05 | May 2026
p-ISSN: 2395-0072
www.irjet.net
ENHANCING URBAN PARKING EFFICIENCY THROUGH MACHINE LEARNING MODEL INTEGRATION C.Nanda Kumar Reddy 1, Naga Sravani 2 student, Mca 2nd Year Kmmips, Tirupati, Affiliated To S.V. University, Tirupati, A.P, India Associate Professor, Dept Of Mca, Kmmips, Tirupati, Affiliated To S.V. University, Tirupati, A.P, India ----------------------------------------------------------------------***--------------------------------------------------------------------1
2
ABSTRACT - An increase in vehicular traffic and a
accuracy of forecasting parking demand, maximized the efficiency of space usage, and provided motorists with rely able information. This, in turn, reduces traffic congestion, enhances the satisfaction of the users, and boosts the overall operational efficacy.
scarcity of parking spaces are creating significant challenges for urban parking management. This study aims to tackle these issues that escalate congestion and pollution and decrease urban productivity, by utilizing machine learning models to accurately predict parking space availability and categorize occupancy levels.
1.1. BENEFITS OF PREDICTION MODELS
Key Words: Parking management, predictive modelling, random forest model, decision tree, support vector machine, linear regression
It is becoming progressively difficult to manage parking in urban settings due to the ever-increasing number of vehicles on the road and the constrained availability of parking spaces. Vehicle ownership has surged with economic growth and urban expansion, intensifying the imbalance between the availability and demand for parking Drivers typically spend from 3.5 to 14 minutes looking for parking spaces, depending on location and time, which induces stress and delays, increases pollution, and wastes fuel Reducing the amount of time spent searching for parking spots can diminish the circulation of vehicles in parking lots, thereby lowering traffic congestion and noise pollution around the entrances. The inefficiency in parking searches stems from the absence of real-time information on space availability. Providing drivers with access to realtime parking maps that reveal available spaces via their vehicle’s navigation systems would not only benefit drivers but also aid parking facility operators and urban planners in managing their space more effectively, based on anticipated demand. In the past, methods for predicting parking space occupancy were primarily based on static guidelines and heuristic approaches, which lacked accuracy and flexibility. These methods typically depended on direct observation, a near row scope of past data, or elementary statistical techniques, which are not adequate for the intricate nature of city parking dynamics. However, the emergence advanced technologies for data collection, the accessibility to wide-ranging parking data sets, and the incorporation of machine learning into these processes have marked a significant move towards approaches cantered around data. According to and, this shift has profound implications. Integrating machine learning into parking management has significantly improved the
|
Impact Factor value: 8.315
AVAILABILITY
Research indicates that the availability of parking spaces significantly influences drivers’ decisions regarding where to park, and a lack of information about available parking can lead to increased time spent searching, cruising, and queuing, which is also costly. For example, in the US, drivers annually spend about 17 hours searching for parking, incur ring a cost of around $345 per person; in the U.K. and Germany, these figures are even higher. Conversely, drivers who have access to information on parking availability are less frustrated and are 45% more likely to make effective parking decisions which reduces traffic and energy consumption and enhances transportation management Intelligent parking management systems and related research have shown that providing parking availability information enriches the customer’s experience and enables online parking reservations. An example is SF park, which was initiated in San Francisco in April 2011 and effectively manages onand off-street parking availability, using intel legend parking meters that adjust pricing based on factors like location, time, and day and aim to keep about 15% of spaces vacant on each city block. The system is equipped with sensors that track space availability and allow users to view prices via SFpark.org and mobile apps. Spark has shown notable results: a 4% reduction in metered parking rates and 12% in city garages, a 43% decrease in time spent searching for parking, and a 30% reduction in daily vehicle miles travelled, leading to safer streets, less con gestion, and reduced neighbourhood pollution. Additionally, Spark has positively impacted local businesses, which have experienced a more than 35% increase in sales tax revenue compared to less than 20% in other city areas. Spark’s achievements underscore the potential of predicting parking occupancy to tackle urban congestion and promote a more efficient, sustainable transportation system. Its use of sophism dictated technology to predict parking availability and adjust prices
I. INTRODUCTION
© 2026, IRJET
PARKING
|
ISO 9001:2008 Certified Journal
|
Page 476