This projectpresents a Smart Parking System SPS designed to address the challenges associated with tra-ditional parking methods and enhance the overall parking experience for both driver
Background and context
Aims and objectives
The primary aim of researching the mobile application component of the Smart Parking System (SPS) is to thoroughly assess and enhance the performance and utility of the existing SPS web application The study focuses on optimizing parking space utilization by leveraging real-time occupancy monitoring and advanced analytics to allocate parking resources efficiently, reducing underutilized areas and overcrowding It also addresses urban mobility by cutting traffic congestion through real-time updates on parking availability delivered via the mobile app and digital signage, enabling drivers to quickly locate vacant spaces and reduce unnecessary vehicle circulation.
Enhancing the user experience is at the core of SPS, delivering a seamless and enjoyable parking journey through features like mobile app-based reservations, cashless payments, and real-time navigation to available spaces These capabilities streamline the parking process, reduce common frustrations, and create a more user-friendly experience for all users.
Data-driven decision making is essential for effective parking management and urban planning SPS collects valuable data on parking demand, space occupancy, and user behavior, and data analytics turn this information into actionable insights By translating these insights into strategic choices—such as infrastructure upgrades, pricing strategies, and policy updates—parking operators and city planners can improve urban mobility and optimize parking resources.
SPS drives revenue generation for parking operators by optimizing space utilization, deploying dynamic pricing strategies, and delivering value-added services that expand earning potential for parking facilities, all while maintaining transparent and equitable pricing for users.
Improving Enforcement and Security: Improving enforcement capabilities and ensuring the security of parking facilities are also critical objectives of SPS By automating the detection of
Real-time alerts sent to authorities help maintain order within parking areas, deter illegal park- ing, and enhance overall safety for users and vehicles.
Through these objectives, the research aims to advance the SPS mobile application into a comprehensive tool that not only improves the efficiency of parking systems but also contributes to the broader goals of sustainable urban mobility and enhanced user satisfaction The scope of this study encompasses the development, implementation, and evaluation of the SPS app as an integrated solution for smarter parking, focusing on efficiency, sustainability, and an improved user experience.
The scope of a Smart Parking System (SPS) encompasses a broad range of aspects that address urban parking management, technological integration, and user experience enhancements It brings together real-time occupancy sensing, data analytics, and seamless payment and reservation capabilities to optimize space use and simplify parking for drivers Within this scope, key components include sensing infrastructure (cameras, meters, and ground sensors), robust communication networks, data processing and analytics platforms, and driver-facing interfaces such as mobile apps and web portals, along with integrated guidance, enforcement support, and feedback mechanisms that benefit operators, city authorities, and users alike.
The Smart Parking System (SPS) focuses on maximizing the efficiency of parking space use by deploying IoT devices—sensors and cameras—for real-time monitoring of parking occupancy and by building a robust hardware and software infrastructure to accurately collect, process, and analyze data on space usage By applying advanced data analytics and machine learning, SPS predicts parking demand patterns, providing predictive insights that support proactive management and enable the efficient allocation of parking resources while minimizing the risk of over- or underutilization.
Thirdly, the SPS features intuitive user interfaces accessible through mobile applications and web platforms, delivering real-time parking availability updates, easy reservation capabilities, and cashless payment options By prioritizing a seamless user experience across devices, these interfaces enhance satisfaction and ease of use for all users.
Designed to integrate seamlessly with urban transportation infrastructure, the system connects with parking meters and public transit networks to enable true interoperability across components This connectivity ensures smooth data exchange, coordinated operations, and smarter parking and transportation management strategies for cities.
Smart Parking System (SPS) provides a flexible, scalable solution designed to support diverse urban environments, accommodate varying parking demand, and adapt to future technological advancements Its modular design enables scalable deployment, ensuring the system can evolve with changing user needs and infrastructure across different cities and regions.
Sustainability is a core pillar of the SPS, driving eco-friendly urban mobility By optimizing parking space utilization, reducing the time spent searching for parking, and promoting alternative transportation modes, the system minimizes vehicle emissions These efforts contribute to greener, more sustainable urban environments.
By leveraging data-driven insights and real-time information, the SPS enhances the management of parking facilities The system enables dynamic pricing, stronger enforcement of parking rules, and optimized resource allocation, collectively boosting efficiency and operational effectiveness across all facilities.
The holistic approach of the SPS positions it as a transformative solution in urban mobility,addressing the multifaceted challenges of parking management while promoting environmental sustainability and user convenience.
Our advancements
Technological advances have transformed many industries, with transportation at the forefront Traditional parking systems, which relied on manual processes, are being replaced by smart parking solutions that utilize cutting-edge technologies These modern systems address the inefficiencies of conventional methods, delivering improved accuracy, greater convenience, and more sustainable urban parking management By optimizing capacity, reducing search times, and enabling data-driven operations, smart parking innovations help cities run more efficiently and eco-friendly.
Key Advancements in Smart Parking System
IoT integration is a cornerstone of modern smart parking systems (SPS) By tying together Internet of Things devices—sensors, cameras, and actuators—these systems enable real-time monitoring of parking space occupancy, automate enforcement processes, and elevate the overall user experience This approach facilitates seamless data collection and analysis, equipping parking systems with efficient, responsive decision-making capabilities.
Artificial Intelligence (AI) and Machine Learning (ML) have transformed smart parking systems by enabling predictive analytics, pattern recognition, and anomaly detection AI-powered solutions forecast parking demand and optimize the allocation of parking resources, while automating tasks such as license plate recognition and rule enforcement These capabilities boost operational efficiency, reduce manual intervention, and improve user convenience, delivering smarter, more responsive parking experiences.
Through data analytics and big data technologies, SPS derives actionable insights from vast volumes of parking-related information By examining historical trends, traffic patterns, and user behavior, SPS optimizes parking operations, forecasts future demand, and informs urban planning These insights support the creation of more sustainable, adaptable, and resilient urban environments.
The widespread adoption of smartphones and digital platforms has enabled the development of user-friendly mobile applications for smart parking systems (SPS) These apps provide real-time parking availability updates, navigation assistance, reservation capabilities, and contactless payment options Together, these features streamline the parking process, boost user satisfaction, and help reduce traffic congestion.
High-performance systems, powered by cutting-edge computing and cloud infrastructure, boost the efficiency and scalability of Smart Parking Systems (SPS), enabling faster data processing and more reliable operations These technologies also deliver cost-effective solutions for managing parking resources across diverse urban settings.
The integration of advanced technologies has elevated Smart Parking Systems (SPS), making them indispensable tools for modern urban mobility By addressing the shortcomings of traditional parking, SPS boosts efficiency, sustainability, and user satisfaction while reducing congestion and improving on-street flow This momentum paves the way for smarter, more connected cities where parking integrates seamlessly with transit networks and digital services.
Description of the Remaining Chapters
In this specialized project, there will be seven main chapters, a section for listing results and future developments and a section for references:
Defining and documenting Introduce some important knowledge and technologies we will use in this system.
Defining functional and non-functional specifications, this part establishes a clear road map for development.
Detailed specifications of use cases offer a comprehensive view of the various use cases within an Smart Parking System.
Chapter 5: System analysis and design
Analyzing user need and designing overall system architecture of entire system to meet business needs through technology.
Chapter 6: Schedule for next phase
This chapter will include our plans for the graduation thesis.
Concluding work that has been done in this specialized project and planning for the next phase.
This paper presents a literature survey of smart parking solutions, highlighting smart sensors, tools, and applications used to improve parking efficiency Since the early 1980s, research has focused on sensor-based and technology-driven approaches, and this review centers on the devices and methods deployed to solve the problem rather than on broader parking requirements or the overall progress of individual studies Parking meters and manual, human-resource–based methods are excluded The evaluation of smart parking tools emphasizes compatibility with outdoor environments and cost considerations, where compatibility is affected by light conditions and environmental changes, and costs cover installation, procurement, and ongoing maintenance.
Mimbela and Klein's 2000 study provides a thorough overview of vehicle detection and surveillance technologies used in intelligent transportation systems (ITS) It surveys the range of methods for detecting and monitoring vehicles within ITS, including sensor-based systems, video surveillance, and other approaches designed to enhance traffic efficiency and road safety.
Idris M., Leng Y.Y., Tamil E.M., and colleagues’ 2009 study provides a comprehensive review of the Sag Park system, focusing on smart parking technologies It surveys the key elements of smart parking systems—from technological components and integration approaches to implementation strategies—and outlines the potential benefits these systems offer, such as improved parking efficiency and enhanced user convenience.
A comprehensive survey by Revathi G and Dhulipala V.S reviews smart parking systems and sensing technologies, detailing sensor options, system architectures, deployment strategies, and their impact on urban mobility and parking efficiency.
Mahmud S., Khan G.M., Rahman M., and colleagues conducted a 2013 survey on intelligent car parking systems, published in the Journal of Applied Research Technology The study reviews the technologies employed in intelligent parking, evaluates system design and architectural considerations, addresses operational challenges, and outlines the potential benefits for urban mobility and transportation management.
Fraifer M and Fernström M examine smart parking systems and their associated technologies, surveying the current landscape of smart parking solutions and highlighting the latest technological advancements The paper discusses implementation challenges and outlines potential applications within IoT-enabled smart city development, illustrating how these systems can integrate into broader urban tech ecosystems.
A comprehensive survey by Hassoune K., Dachry W., Moutaouakkil F., et al investigates smart parking systems, detailing how sensor networks, data management, and user interfaces support smarter parking and how these elements integrate into intelligent transportation frameworks The review also examines existing approaches to smart parking, comparing strategies, technologies, and systems currently deployed to address urban parking challenges and to improve urban mobility, efficiency, and user experience.
After reviewing the studies cited above, this literature review highlights the general smart tools and technologies that have been used—and are still being used—in parking systems, as shown in the table below However, most of these studies have not focused on parking issues related to open parking spaces.
Table 2.1: Literature reviews on smart parking sensors and technologies
Smart parking sensors and tools
Infrared sensors
Infrared sensors detect changes in energy—such as heat and light—to determine occupancy by measuring how an object like a car or a person alters the local energy field Passive infrared sensors rely on ambient energy changes, but they are highly sensitive to environmental conditions and can become inaccurate in poor weather such as rain, snow, or fog These passive devices are typically installed underground or on ceilings and require significant setup and maintenance costs, which limits their use to closed parking spaces The article then introduces active infrared sensors, which use emitted signals to detect occupancy and offer different benefits compared with passive systems.
Instead of detecting the changes of energy, active infrared sensors emit infrared energy and if there are objects above them, the sensors can detect them by the reflected energy
Active sensors differ from passive sensors in their detection method, but their overall effectiveness is comparable As a result, they are well-suited for closed parking spaces where environmental conditions have minimal impact, and they are not suitable for open parking spaces.
Ultrasonic sensors
Ultrasonic sensors operating in the 25–50 kHz range emit and receive reflected energy to detect and distinguish vehicles and pedestrians by distance, a capability well suited to indoor parking environments Because these sensors are environment-sensitive, they are typically ceiling-mounted in closed parking spaces to minimize interference They come in a range of low-cost options, but the total cost of installation, integration, and ongoing maintenance can be high These sensors usually connect over wireless networks using ZigBee or similar protocols, enabling scalable, low-power communication across the system.
To accurately determine parking space availability, ultrasonic sensors should be installed at the top of each space In some surveys, sensors are mounted on vehicles and data are collected periodically, but this approach cannot provide accurate real-time parking data for the lots.
Inductive loop detectors
Underground magnetic detectors detect vehicles by monitoring changes in magnetic fields through a wired system installed at parking lot entrances and exits, providing real-time vehicle counting and occupancy data These magnetic parking detectors are widely used in both closed and open parking areas of shopping malls to track inflow and outflow and to indicate whether parking spaces are available They deliver accurate overall parking occupancy in real time, helping manage space utilization and reduce congestion However, they do not reveal the availability of individual parking slots, and the cost to install and maintain these systems can be high.
Parking guidance systems
Parking guidance systems are a key component of smart parking, equipped with display screens that show the number of available and occupied parking spaces and are typically placed at the entrances of different parking zones to help drivers quickly decide where to park While they provide an overall occupancy view, these systems do not offer real-time availability for each individual parking spot or step-by-step guidance to reach a free space, so drivers may still end up driving around the area to locate an empty slot.
The parking management system integrates detection tools such as inductive loop detectors and visual cameras to accurately monitor and regulate the number of vehicles occupying parking spaces These sensors are specifically deployed to control vehicle counts, delivering a cost-effective solution that keeps overall costs low This approach is suitable for both open and closed parking spaces, making it practical for a wide range of parking facilities.
Radio frequency tags
Radio frequency identification (RFID) tags identify each vehicle as it enters and exits a parking facility, with a transceiver and antenna at the gateway to control access and manage parking flow In this setup, RFID serves for in/out vehicle access control rather than detecting the availability of individual parking slots or guiding drivers to vacant spaces As a result, RFID is best suited for closed parking areas with a registered, limited number of vehicles and is not appropriate for open-space parking where slot availability and guidance to empty spaces are needed.
Magnetometer
Magnetometer-based parking sensors detect vehicles by monitoring changes in the electromagnetic field To operate effectively, each sensor must be placed in close proximity to its corresponding slot, typically installed underground within the slot They are robust to environmental changes, and some wireless options offer multi-year battery life, making them well-suited for open parking areas With a magnetometer installed at every slot, the system provides real-time parking space availability While installation and maintenance costs are reasonable per slot, outfitting large facilities with hundreds or thousands of spaces can be expensive because every slot requires its own sensor.
Microwave radar
A single microwave radar emits a microwave beam and analyzes the reflected signal to estimate parking occupancy, and it can be upgraded to a dual microwave Doppler radar to reliably detect both moving and stationary vehicles With Doppler processing, the system can distinguish vehicle motion and provide accurate occupancy data for a range of parking environments Because it shares magnetometer-like attributes, the radar works in both open and closed parking spaces, but scaling the technology for large deployments requires careful cost considerations.
Smart parking technology
Global positioning system (GPS)
Global Positioning Systems (GPS) are widely used to guide drivers to destinations, optimize routes from the current location to the nearest parking spaces, and help users locate an available parking spot However, GPS provides only navigation and cannot supply real-time parking availability To address this limitation, a study integrates historical parking-space data into GPS to estimate parking space availability [17].
GPS works by emitting and receiving signals from satellites, delivering accuracies of under 7.8 meters with a single-frequency receiver and under 0.71 meters with a dual-frequency receiver A typical parking space measures about 2.3 to 2.7 meters in width However, GPS signals can be blocked or degraded by obstructions such as tall buildings, walls, or ground in underground parking, leading to errors Consequently, GPS performs best in open outdoor spaces rather than closed indoor parking lots, and satellite availability also impacts GPS signal quality.
Machine vision
License plate recognition with visual cameras and computer vision can monitor vehicle flow by placing a gateway camera to count entries and exits, providing insight into the number of available parking spaces Yet inspecting every individual slot is unreliable with this method, since a wide field of view can obscure some plates and continuous video processing of each space demands substantial bandwidth Therefore, machine vision is better suited for slot-number control and empty-slot detection To detect empties, an overhead camera positioned above the spaces, paired with a suitable computer-vision model, analyzes each slot; the video stream can be divided into periodic images to reduce bandwidth pressure.
Open-source datasets like PKLot offer a vast collection of outdoor parking space images for research In studies of parking-slot detection, a support vector machine (SVM) classifier achieved 89% accuracy on the test data, while other methods such as logistic regression and the Viola–Jones detector have shown equal or better performance.
Visual cameras are a strong choice for outdoor parking spaces because they capitalize on natural light and provide a wide field of view that can cover large portions of the parking area Although glare, shadows, adverse weather, occlusion, and lighting changes can reduce accuracy, these issues can be mitigated by leveraging 3D scene information In addition, the ability of a single camera to monitor a large part of the parking space helps keep deployment costs low.
Vehicular ad hoc networks (VANET)
This smart parking system leverages a network of connected wireless devices to deliver real-time guidance and anti-theft capabilities across parking facilities It uses road-side units (RSUs) deployed throughout parking lots and individual spaces, along with on-board units (OBUs) installed in participating vehicles and managed by a trusted third party As a vehicle approaches RSUs, up-to-date parking-space information is transmitted to its OBU, enabling drivers to locate available slots efficiently The same mechanism also supports autonomous driving applications and enhances collision avoidance by providing accurate parking data.
RSUs and OBUs are designed to be environment-insensitive, making VANET-based parking solutions suitable for both open and closed spaces For accurate parking occupancy data and reliable navigation information, all vehicles in the network must install OBUs, since data quality suffers when vehicles without OBUs enter the parking area Consequently, deploying Vehicular Ad Hoc Networks (VANETs) entails high installation and ongoing maintenance costs.
Multi-agent systems
This smart parking system combines sensors, mobile devices, intelligent algorithms, and visual cameras into an integrated solution that streamlines finding an empty parking slot Its integration capability allows it to account for user preferences and priorities, using real-time data to optimize the search and quickly identify available parking spaces.
This system forms the foundation for smart, automated parking, typically deployed on web or mobile applications with versatile UX/UI Java-based development tools like JaCaMo and the CARTAGO environment can be integrated into the architecture to support multi-agent coordination In addition, multi-agent systems offer high compatibility and can leverage alternatives such as VANETs or machine vision to replace traditional sensors Consequently, multi-agent–based parking solutions are suitable for both open and closed parking spaces, with costs depending on the tools and technologies selected.
Neural network
Neural networks are a foundational concept in artificial intelligence (AI) and deep learning, inspired by the brain’s neural structure Over the years they have evolved into diverse architectures, including fuzzy, fluid, feed-forward, and convolutional neural networks They are widely used in computer vision to enable smart automation, notably in license plate recognition A study used day and night images to train a two-layer feed-forward network with a sigmoid hidden layer, achieving highly accurate detection of available parking spaces.
Deep learning, a branch of machine learning, uses neural networks to detect and classify objects in images By feeding images into convolutional neural networks, the models improve their analysis as they learn In computer vision, these networks can efficiently determine the availability of parking spaces by recognizing empty spots in real time.
Convolutional neural networks (CNNs) are increasingly used in autonomous driving, leveraging camera-captured images to train perception and decision-making systems With limited training data, vehicles equipped with CNNs can operate on roadways and in parking areas even when lane markings are absent This capability expands where self-driving cars can safely navigate, from cluttered urban streets to parking lots, by enabling robust scene understanding and driving control from camera inputs.
This data processing technology remains effective when real-time data capture is not involved, delivering reliable performance without the overhead of live data streams Consequently, it offers a cost-efficient solution for both closed and open parking spaces, providing flexible applicability across different parking facility setups.
Fuzzy logic
Fuzzy logic uses multivalued logic for evaluation and, like neural networks, can be used to develop prediction models from sample data and integrate them into multi-agent systems For example, a dataset of supermarket parking information collected over five days can train a fuzzy-logic-based model to predict parking space availability, enabling more accurate forecasts of when spaces will be empty.
Fuzzy logic aids drivers in decision making when searching for a parking slot, providing guidance under uncertainty about slot availability Yet the accuracy of such an approach remains limited unless it is continually compared with and evaluated against a real-time parking database By integrating real-time data, the system can refine recommendations and improve reliability as conditions change.
Integrating fuzzy logic with computer vision and sensing technologies delivers real-time parking data and decision-making for both open and closed parking spaces at a cost-effective price This approach enables accurate occupancy detection, reduces driver search time, and optimizes space utilization across facilities with minimal expenditure The fuzzy logic layer handles uncertainty in lighting and environmental conditions, while computer vision and sensors provide robust inputs, creating a scalable, low-cost parking management solution.
Current research indicates that not all existing smart tools and technologies are suitable for a smart parking system that requires real-time availability information Although sensors are widely used in IoT, detecting and collecting parking-space availability data at scale remains costly, and many sensors are too sensitive to environmental changes to perform reliably in open parking areas However, viable alternatives exist, such as machine vision, multi-agent systems, and GPS, which, when effectively leveraged, can meet the real-time needs for both open and closed parking spaces Efficiently exploiting these technologies offers a path toward a robust, scalable smart parking solution that balances accuracy, cost, and resilience The accompanying assessment of the strengths and weaknesses of these tools provides a clearer basis for design choices and deployment strategy.
Table 2.2: Strengths and Weakness of smart parking tools and technologies summary
Suitable for open parking space?
Active/Passive infrared sensor No Provides parking occupancy status
Not suitable for open parking lot due to varying environmental conditions
Ultrasonic sensor No Provides parking occupancy status
Not suitable for open parking lot due to varying environmental conditions
Facilitates in getting the count of vehicles in a parking lot
Not suitable for individual parking occupancy status
Guide driver to a parking lot based on count of available parking spaces
Do not solve congestion while cruising for parking space
Facilitates in authorising vehicles in a parking lot
Not suitable for open parking lots and cannot provide parking occupancy information
Magnetometer Yes Provides parking occupancy status
Expensive to install and maintain on large scale
Microwave radar Yes Provides parking occupancy status
Expensive to install and maintain on large scale
Provides navigational directions to the parking space
Does not provide parking occupancy status and is subjective to errors Machine vision Yes
Provides parking occupancy status or facilitates in authorising vehicles
Challenges with low lighting conditions
VANET Yes Provides parking occupancy status
All vehicles should be part of the network without which will be prone to errors
Multi-agent systems Yes Provides parking occupancy status
Challenges will be based on the
Facilitates efficient parking occupancy detection
Is used only as a supporting tool in parking occupancy detection
Provides parking occupancy status based on historical data
Do not provide real-time parking occupancy
Smart parking applications
My Parking
My Parking is a smart parking service developed by Viettel that connects mobile apps and traffic systems to guide users to available parking locations and enable quick, convenient reservations It provides real-time information on available spots and the status of parking lots, and it offers the optimal walking-distance route to the chosen location, helping drivers save time and reduce congestion.
Effortlessly search for parking with real-time availability of vacant spaces and transparent pricing, then complete a quick, intuitive booking Our platform provides real-time parking information, straightforward booking steps, and clear price insights, so you can reserve a space in just a few clicks You can also search and review your booking history, making it easy to track past reservations and revisit ticket details whenever needed.
– Payment: Supports a variety of modern payment methods including e-wallet
MoMo, VNPay, ZaloPay, ATM/Visa/Master card, ViettelPay, electronic payment gateway
Parking management software provides a centralized view of all vehicles parked in the lot, including the parking time for each vehicle, enabling parking supervisors to monitor occupancy, enforce rules, and quickly issue violations on the go using mobile devices instead of bulky paperwork The system delivers clear visibility into violations across parking lots under management, streamlining enforcement, auditing, and reporting while improving accuracy and response times.
Parking lot administration module enables decentralized control and streamlined system management It lets users access comprehensive information—from equipment installed at the parking facility and general parking lot details to revenue data and recorded violations—and it supports exporting statistical reports for sharing and analysis Strengths include improved accessibility, transparent governance, and the ability to generate actionable insights through exportable data.
– Real application:My Parking has been successfully implemented across various districts, establishing its practicality and effectiveness in the Vietnamese market.
– Unique offering: As the pioneer in the Vietnamese market, My Parking brings a unique solution to urban parking challenges.
My Parking offers a promising solution for urban parking management For continued success and growth in today’s dynamic urban landscape, it will be essential to expand parking availability, streamline processes, and enhance technological features.
EasyPark
Since 2001, EasyPark has been dedicated to enhancing urban living Trusted by millions of drivers, businesses, and operators across over 20 countries, the company continually innovates to offer convenient, user-friendly parking solutions that save time, save money, and remove the stress of car parking.
As Europe’s leading parking app, EasyPark provides wide coverage and a seamless mobile solution that lets users effortlessly pay for parking in garages or locate street parking in city centers, airports, and even abroad Wherever their journey takes them, EasyPark delivers a smooth parking experience In addition to parking, the platform enables electric vehicle charging, adding convenience and supporting sustainability.
– Find parking near your location or destination before traveling.
– Charge electric car with the same app.
– Pay for parking, whether privately or for work.
– Pay with secure methods such as Visa, Mastercard, PayPal, Google Pay, or monthly invoices for business accounts.
– Get notified when users parking is close to expiration. c Strengths
Since 2001, EasyPark has revolutionized parking with user-friendly solutions trusted by millions across 20 countries It offers seamless features such as quick location searches, electric vehicle charging, and secure payment options, delivering stress-free parking experiences Committed to enhancing urban living through ongoing convenience and innovation, EasyPark remains at the forefront of smarter parking.
DFPS - A Distributed Mobile System For Free Parking Assignment [31] 16
This free, peer-to-peer parking assignment system enables users to find and allocate spaces without fees, offering a cost-effective solution that does not rely on any sensing infrastructure It comprises two main components: a mobile app running on users’ smartphones and a server-based dispatcher that receives parking requests and forwards them to other users who can provide spaces By removing the need for sensing hardware, the system lowers deployment costs while supporting real-time, crowdsourced parking assignments This architecture combines user-driven matching with centralized request routing to create scalable and flexible urban parking coordination.
* Peer-to-Peer Parking Assignment: Utilizes a peer-to-peer network for free park- ing space assignment.
* Privacy Protection: Allows users to set privacy levels for parking requests, en- suring anonymity.
* Cloaked Region Generation: Computes cloaked regions to conceal real destina- tions while requesting parking.
* Ticket System: Users earn ’tickets’ by participating in parking assignment tasks, ensuring active involvement.
* Overlay Network: Utilizes a K-D tree network to organize parked drivers and managed regions efficiently.
* Parking Space Assignment Algorithm: Optimizes parking space assignments based on travel time considerations.
* Periodic Parking Space Allocation: Adjusts parking space allocation based on request urgency and availability. c Strengths
* Scalability and Performance: DFPS employs a peer-to-peer network for parking space assignment, reducing reliance on centralized infrastructure and promot- ing community-driven solutions.
Privacy protection is strengthened by allowing users to set privacy levels for parking requests, ensuring anonymity and safeguarding sensitive location data The system generates cloaked regions that conceal the real destinations while requesting parking, thereby enhancing user privacy without sacrificing functionality.
* Active User Involvement: The ticket system incentivizes user participation in parking assignment tasks, fostering a sense of community engagement and re- sponsibility.
* Optimized Parking Space Assignment: DFPS employs an algorithm to optimize parking space assignments based on travel time considerations, enhancing user experience and efficiency. d Weaknesses
The parking system hinges on active user participation, requiring users to perform parking assignment tasks to earn tickets This reliance on user involvement can lead to lower engagement and potential dissatisfaction among users who prefer a more passive experience with the system.
User complexity increases when users must understand and adjust privacy settings, participate in parking assignment tasks, and manage tickets, adding cognitive load that can degrade the user experience and deter some people from using the system To improve usability and adoption, streamline privacy controls, simplify parking-related tasks, and automate or clearly label ticket management to reduce friction and boost overall satisfaction.
Resource limitations in DFPS stem from its reliance on a peer-to-peer network of parked drivers to manage parking assignments This setup can lead to constraints, particularly in areas with low user adoption or during peak demand periods Consequently, the availability and efficiency of parking space allocation may be affected.
DFPS effectiveness hinges on widespread driver adoption to ensure sufficient participation and the availability of parking spaces When coverage is limited or adoption is low in certain areas, the system's overall effectiveness and utility may be diminished.
DFPS delivers an innovative decentralized parking allocation system with strong privacy features and proven scalability Yet reliance on user participation, inherent complexity, and scalability hurdles could hamper practical adoption and prevent it from meeting the requirements of a true Smart Parking system Still, we can translate some strengths from this model into our solution, notably the K-D tree network approach and a privacy-focused stance on user location, which represent valuable directions for building a more effective smart parking platform.
Research gap discussion
Today, most existing smart parking apps rely on a range of sensors to monitor and report the availability of parking slots in closed facilities, which entails high installation and maintenance costs; for open parking spaces, offerings exist mainly as basic services—listing parking locations and predicting availability—yet they remain imperfect and limited in scalability Apps like AppyParking can show the locations of available spaces but not the actual occupancy, while Viettel’s MyParking can provide both location and occupancy data, albeit with limited coverage since it only includes spots connected to and managed by Viettel Reservation and authorization features tend to be weak and overly dependent on human staff A handful of platforms, such as EasyPark and Parkopedia, use predictive analytics or crowdsourcing to address open-space occupancy, and such analytics are typically cheaper than sensor deployments and therefore common in many cities, but the occupancy information from predictive analytics and crowdsourcing is not always reliable or complete.
Despite the promise of “real-time” data, none of the apps currently available provide live parking availability for open spaces Theoretically, a combination of smart tools and technologies could yield real-time information about open parking, yet no country or organization has implemented a full solution One reason may be the lack of economic returns, since open parking spaces are typically outdoor and free, which leads people to assume high costs and low profitability Consequently, there are no immediate financial gains from deploying smart parking tools for open parking spaces.
Smart parking applications function as decision-support tools that help ease traffic congestion by guiding drivers to available parking spaces However, adoption is optional, and some drivers remain comfortable with traditional parking methods, as illustrated by Viettel's MyParking app Real-time occupancy information improves efficiency only when nearly all drivers participate, a scenario that is unlikely in practice.
According to a prior review, while numerous smart sensors, tools, and technologies exist, none can achieve efficient parking in open parking spaces Infrared and ultrasonic sensors are highly sensitive to environmental changes, whereas magnetometers and radars remain robust to such fluctuations but are often prohibitively expensive As a result, sensor-based solutions are not ideal for open parking environments, leaving the problem to alternative tools and technologies for improving parking efficiency in outdoor lots.
Vehicular Ad Hoc Network (VANET) is a promising approach for enhancing traffic safety and efficiency, but its full potential hinges on widespread adoption of on-board units (OBUs) in vehicles This universal OBU deployment is more feasible in the future than today In addition, the deployment and installation of roadside units (RSUs) involve significant costs and construction challenges.
Addressing the challenges of light and shadow, machine vision integrated into a visual camera with neural networks provides a feasible solution for real-time parking occupancy information in both open and closed parking spaces Neural networks, including convolutional neural networks (CNNs) and other deep learning algorithms, are proven efficient for image detection and classification Although CNNs can achieve high image classification accuracy, there are relatively few studies applying convolutional neural networks to parking occupancy detection, with most research using broader deep learning approaches.
Deep learning leverages large public datasets and benchmarks such as PKLot, and architectures like AlexNet, to advance image classification Considering cost efficiency, the combination of machine vision, multi-agent systems, and neural networks presents a sensor-free solution for occupancy detection in open parking spaces, enabling accurate vision-based parking monitoring without additional sensors and reducing hardware expenditure while supporting scalable smart parking applications.
Machine vision can cover a wide field of view per camera, potentially reducing the total number of cameras needed for parking occupancy detection A range of algorithms—logistic regression, radial basis function, support vector machines, decision trees, random forests, Viola-Jones object detectors, and others—are suitable for pattern recognition and image classification tasks in this context The use of visual cameras to collect occupancy data is governed by privacy laws that vary by country: in the UK and the USA, deployments in public spaces face fewer restrictions, whereas in Sweden surveillance is restricted unless for crime prevention or investigation Surveillance cameras can be used without constraints in private areas not accessible to the public In Sweden, open parking-area surveillance is regulated by privacy laws, so camera usage may be subject to national regulations To stay compliant, thermal cameras are suggested because they do not publicly identify individuals While thermal cameras cost more than color cameras, they can cover many parking spaces—often more economically than underground sensors—and future research could evaluate their effectiveness for parking occupancy detection.
An integrated neural network can determine the availability of parking slots, and for convolutional neural networks (CNNs), building an effective model requires collecting representative train and test images The accuracy of parking slot classification largely depends on the quality and diversity of these training samples, which should include various vehicles and parking spaces under different environmental and lighting conditions A comprehensive dataset that captures such variation is essential for training CNNs to achieve reliable and efficient image classification and accurate parking occupancy detection.
All processes are integrated within a multi-agent system that visualizes data through applications and offers a navigation service The smart parking application supports decision-making by helping users locate parking spaces and slots; however, real-time updates of availability remain essential, because a selected space may be occupied by other drivers whether or not they are registered in the system—a challenge that persists for open parking spaces without sensors.
Therefore, the short-term prediction of occupancy rises as a potential method Because the reservation term is not practical and possible for open parking spaces, algorithms like fuzzy logic and time series used for predicting short-term can help in real-time update problems Parking space availability can be forecasted by considering the driver’s location and the distance to their intended parking spot or destination By determining an optimal route based on this distance, an estimated arrival time can be calculated Consequently, the driver can be provided with predictions of parking space occupancy based on this estimated arrival time Through this process, smart parking applications can function as effective decision-making tools.
Overall, automation is the trend nowadays, specifically for the parking system is the park assist integrated in some smart vehicles, which help drivers to park automatically but with an insight specific parking spot and not help much with traffic congestion But imagine that there is real-time parking availability information put into the park assist, all the ve- hicles now can auto park without any need of human intervention Therefore, the choice and integration of multiple smart tools and technologies to address the problem of smart parking not only helps for the traffic and driver convenience but it is also the foundation of the development of smart cities of automation and convenience, further smart country, world The choice of smart parking technology or sensors for gathering parking occu- pancy data can vary depending on the type of parking lot For indoor parking lots with a limited number of spaces, sensors like magnetometers can be employed effectively Con- versely, for outdoor parking lots, visual cameras positioned at an elevated vantage point can be utilized to monitor the entire area Therefore, one of the most important things we need to do now is determine the specific research gap to exploit, develop and which tools/technologies to use in the upcoming our own smart parking application.
Conclusion
Although there are substantial research gaps and a wide range of smart tools and technologies to choose from, any solution must consider cost-effectiveness, scalability, and the feasibility of integrating with existing infrastructure Regulatory compliance is particularly important in Vietnam, a developing country with limited infrastructure and relatively low car ownership, where parking regulations are still incomplete and not yet fully clear The goal should extend beyond academic growth to actively contribute to the development of urban mobility solutions that address real-world parking and traffic management challenges.
To meet the distinct needs of open and closed parking spaces, the feature set should be strategically tailored to each space's function Drawing on prior research, the capstone project aims to advance Smart Parking Systems (SPS) by prioritizing predictive analytics, sophisticated mapping and routing, and enhanced computer vision techniques These innovations will address current gaps and anticipate future demands in urban parking management.
Predictive analytics is a cornerstone of this project, using neural networks and fuzzy logic to generate both short-term and long-term parking availability predictions By integrating time-series data, traffic patterns, and historical parking trends, the system delivers actionable forecasts that help drivers plan more effectively Enhanced algorithms adapt to dynamic urban environments, accounting for variations in user behavior and the impact of special events, ensuring robust performance across changing conditions.
Mapping and routing systems will transcend conventional navigation by integrating parking-specific data A dynamic mapping solution will combine real-time occupancy data with predictive insights to guide drivers toward optimal parking zones rather than individual spots The routing approach will prioritize minimizing travel time and fuel use while accounting for traffic flow, parking fees, and user preferences.
In parking management, state-of-the-art computer vision systems will be deployed to improve parking space detection and monitoring in both open lots and enclosed facilities Neural networks trained on diverse datasets will enhance the accuracy of key vision tasks, including occupancy detection, vehicle classification, and license plate recognition These advanced capabilities will streamline parking enforcement and strengthen overall security.
Enhancing user experience through user-centric features such as reservation capabilities, secure access via advanced authentication methods—including machine-vision with RFID—and predictive parking guidance will be implemented Rather than directing drivers to individual spots, the system will guide them to areas with the highest likelihood of availability, improving efficiency, reducing search time, and easing congestion while supporting data-driven parking management.
In Vietnam’s evolving urban landscape, characterized by limited infrastructure and shifting parking regulations, this project emphasizes sustainable and scalable, cost-effective solutions Its modular design enables seamless integration with existing infrastructure, while predictive insights drive emission reductions and optimized resource utilization, supporting long-term sustainability and adaptability.
By exploring these specialized domains, the project aims to develop innovative, data-driven solutions that enhance urban mobility These efforts contribute to academic research while paving the way for smarter, more sustainable cities.
Requirement analysis and detailed specification of use cases
Building on surveys that identify key stakeholders and gather functional and non-functional requirements, this chapter illustrates the Smart Parking System (SPS) using a use-case diagram that visually presents its operation across diverse real-world scenarios—from the initial step to the end of the order process—and offers stakeholders, developers, and end users valuable insights into how the system simplifies workflows and enhances the customer experience.
Stakeholder
Stakeholders play a pivotal role in the system’s development and operation Understanding their needs and expectations is essential for building a robust SPS The two primary stakeholders identified are:
• Administrator: The administrator is responsible for managing the overall system, includ- ing its operational, security, and user-related aspects Their primary functions include:
– Overseeing parking space operations through detailed mapping and real-time mon- itoring of parking lot statuses.
– Managing parking reservations, user accounts, and system configurations.
– Generating comprehensive reports on system performance, user activities, and rev- enue metrics.
– Implementing and maintaining the security framework to protect user data and en- sure system reliability.
• Driver: drivers are registered customers who have created an account on the platform.
– Searching for and navigating to available parking spaces using a user-friendly mobile application.
– Managing user accounts, and system configurations.
– Completing cashless transactions for parking services and managing their account details.
– Receiving personalized recommendations based on historical parking data and user preferences.
Functional requirement
The system's functional requirements are organized by function to make them easy to understand and use, with each requirement detailing a specific capability the system must perform By categorizing these requirements according to system functions, we provide a clear, user-friendly map of what the system can do and how its features fit together The list highlights individual functional needs—each describing a concrete action the system must execute—so readers grasp the scope, interfaces, and expected behavior of the software.
• Access a detailed, interactive map of parking spaces.
• Monitor real-time parking space occupancy and associated vehicle details, such as license plates and parking durations.
• Modify parking configurations, including adding new spaces, locking/unlocking specific slots, and updating space statuses
• Integrate and manage IoT devices, such as cameras and sensors, for enhanced oper- ational control.
• Manage system accounts, including registration, profile updates, and user access permissions.
• Generate detailed reports on parking occupancy, revenue, and system performance.
• Address user feedback and generate actionable insights.
• Implement real-time alerts for suspicious activities or unauthorized system access.
- Parking Space Detection and Navigation
• Access an interactive map to locate parking spaces and receive navigation assistance to the nearest available spots.
• View detailed parking lot statuses, including real-time availability and demand fore- casts.
- Vehicle Recognition and Authentication (Just for registered users)
• Enable vehicle recognition through license plate authentication.
• Manage vehicle ownership records through the application.
- Account features (Just for register users)
• View and manage payment histories and past parking activities.
• Provide feedback and communicate directly with administrators for support.
• Users’ locations are hidden and all other data also is encrypted before storing
• Implement alarm systems or real-time alerts to suspicious activities or unauthorized access.
• Collect and analyze parking usage data to optimize parking space allocation and improve operational efficiency.
• Generate reports of user demographics, parking activities and other key metrics.
- Provide customer support channels for users
Non-functional requirements will be divided according to different functional blocks of the system, with the three main agents being: user, guest, and shipper.
1 Ensure quick response time for any requests
2 Ensure the scalability in numerous processing requests without errors or crashing as well as the growth of number of users.
3 Ensure numerous requests can be through concurrently without any lost.
1 Ensure the availability of the system services during continuous operation, especially peak hours.
2 Ensure fault tolerance with redundancy and diversity of the system.
3 Ensure CIA of data and security with various encryption methods
1 In the beginning, the system will support 2 languages Vietnamese and English and will expand with more languages in future.
1 Design user-friendly interfaces with clear guidance, icons, buttons,
2 Ensure the app can be compatible with various OSs, displays of different devices.
1 Implement AES-256 encryption standard for data transmission.
2 Users should have secure methods to log in, and their actions should be limited based on their roles and permissions.
Use case diagram
Whole System
Figure 3.1: Driver use case diagram for the smart parking system
Figure 3.2: Parking owner use case diagram for the smart parking system
Table 3.1: General use cases description for the whole system
ID Use case Actors Description
The user log in into system
The user create view/update account pro- file
4 Customer support Driver, parking owner
The system support for the user, and user can feedback or report
The user can register to use app.
6 Live communication Driver, parking owner
Driver and parking owner can communi- cate by live chat
Driver Driver booking parking lots
Driver Drivers can search their vehicle in parking lots, if they forgot
9 Check in/ Check out Driver Offering drivers auto payment options via a mobile application
Parking owner The parking owner can manage their park- ing space
11 Parking space reser- vation management
Parking owner The parking owner can manage their reser- vation request
12 View report Parking owner The parking owner can view parking occu- pancy, revenue,
Table 3.2: Use case scenario for "Login"
Description: This use case is login function
Trigger: User clicks on login button
Precondition: • The user has already registered an account with the system.
• The user is on the login page.
Postcondition: • If the login was successful, the user is now logged in and can ac- cess features and functionalities of the system that require authenti- cation.
• If the login was unsuccessful, the user remains on the login page.
1 The user enters their username and password into the respective fields.
2 The user clicks the “Log In” button.
3 The system validates the entered credentials.
4 If the credentials are valid, the system logs the user in and redirects them to the homepage.
5 If the credentials are invalid, the system displays an error message and prompts the user to try again.
Exceptions: If the user forgets their password, they can click on the “Forgot Pass- word” link to reset their password.
Table 3.3: Use case scenario for "Map"
Description: This use case is for user access map
Precondition: The user is logged into the system.
Postcondition: User can looking for parking space such as nearest space or specific space and then they are navigated to the parking space
1 The user looking for parking space
2 The user looking for nearest parking space
3 The user are navigated to the parking space
Alternative flow: 2a If user don’t want to looking for nearest parking space, they can looking for specific parking space
3a If user don’t want to navigated to the parking space, they can view parking space information
4 The user can looking for available parking lots
5 The user are navigated to available parking lots
Table 3.4: Use case scenario for "Account management"
Description: This use case is for user manage account
Precondition: The user is logged into the system.
Postcondition: The user access account information
Normal flow: 1a The user access profile
2a The user view and edit profile 3a The user view and edit vehicle/parking space ownership
1b The user access Payment information 2b The user view and edit payment information 3b The user view payment history
1c The user access Activity history 2c The user view activity history
Alternative flow: 1d If the user are driver, they can access register vehicle ownership
2d Drivers fill in vehicle information 3d Drivers upload roof
Table 3.5: Use case scenario for "Customer support"
Description: This use case describes how a user can use customer support function
Trigger: The user selects the ‘Customer support’ option
Precondition: The user must be logged into their account.
Postcondition: The user successfully access customer support and choose type of sup- port.
1 The user access customer support 2a The user choose support channel and get help from the system 2b The user submit feedback or report to the system
Table 3.6: Use case scenario for "Register"
Description: This use case describes how a user can register account
Trigger: The user selects the ‘Register’ option
Postcondition: The user successfully access register page.
1 The user choose role to register
2 The user fill in personal information.
3 The user fill in payment information.
4 The user authenticate and complete
5 The application saves the information in the system.
Alternative flow: 4a If the user are parking owners, they must fill in parking information with proof, then authentication and complete
Exceptions: If the user fail in authentication, system will display message please au- thenticate again.
Table 3.7: Use case scenario for "Live communication"
Description: This use case describes how a user can communicate with each other
Trigger: The user selects the ‘Live communication’ option
Precondition: The user must be logged into their account.
Postcondition: The user can communicate together
1 The driver choose parking lots and select live communication.
2 The system connect driver with the parking owner.
3 The user can communicate by live chat
Table 3.8: Use case scenario for "Parking space reservation"
Description: This use case describes how driver can booking a parking lot.
Trigger: The user selects the ’booking’ option with desired parking lots
Precondition: The user must be logged into their account.
Postcondition: The user successfully booking a parking lot.
1 The user logs into the system.
2 The user choose parking space.
3 The user view parking space availability
4 The user choose a parking lot and booking
5 The user fill in information
6 The user make deposit and complete reservation
Alternative flow: 2a If the parking have no availability lot, driver turn back to step 1 and choose another place.
If the deposit is unsuccessfully, system display error message and require driver to deposit again if it still not success system display error message fail to book a parking lot.
If the parking owner reject the request, system send notification to the
Table 3.9: Use case scenario for "Searching vehicle location"
Description: This use case describes how the driver can locate their vehicle in parking space
Precondition: The parking space must have system to navigate.
Postcondition: The driver successfully locate their vehicle.
1 The Driver fill in number plate of their vehicle to the device
1 If the driver fill in plate number that not appear in parking lot, sys- tem display error message.
Table 3.10: Use case scenario for "Checkin/ Check out"
Description: This use case describes how a driver can check in/ check out the parking lots
Precondition: The driver must be logged into their account.
Postcondition: The driver successfully check in/ check out
1 The driver access parking lots
2 Information of check in sent to the system
3 The driver parking their vehicle
4 The system automatic collection fee via app
5 The driver complete check out
1 If there is an error in payment, the driver can not check out until the payment complete.
Table 3.11: Use case scenario for "Parking space management"
Description: This use case describes how a parking owner can manage their parking space
Trigger: The parking owner selects the ‘Parking space management’ option in the system.
Precondition: The parking owner must be logged into their account.
Postcondition: The parking owner successfully manage their parking space
1 The parking owner choose one of their parking space 2a The parking owner access parking space map
3a The parking owner can view number of parking lots, camera 4a System display number of occupied lot
5a The parking owner can view vehicle information, occupied time, occupied fee of customer
2b The parking owner can adjust status lock/ unlock parking space on map
2c The parking owner can view parking space information 3c System display parking space location
4c System navigating parking owner to the address
Alternative flow: 1A The parking owner can adding parking space
2A The parking owner fill in parking space information with proof 3A System require parking owner authentication
4A If authentication is true complete
1B The parking owner can delete parking space [2B.] Parking space removed from system
Exceptions: 2A If the proof is not true, system send reject message to parking owner
Table 3.12: Use case scenario for "Parking space reservation management"
Name: Parking space reservation management
Description: This use case describes how parking owner manage parking space reser- vation
Trigger: The user selects the ‘Parking space reservation management’ option in the system
Precondition: The user must be logged into their account
Postcondition: The user successfully manage parking space reservation
1 The parking owner can manual process request, disable/ enable reservation, and view notifications
2 The parking owner navigates to the ‘manual process request’ sec- tion.
3a The parking owner can assign vehicle 4a The parking owner choose connected driver with their vehicle and assign them to parking lots
5a Driver acceptable 6a System turn to payment processing and complete
3b The parking owner access adjust reservation fee 4b The parking owner fill in the fee
5b Fee save to the system
3c The parking owner access view history reservation activities 4c System display history activities
3d The parking owner can accept/reject request 4d If accept system turn to payment processing and complete
Table 3.13: Use case scenario for "View report"
Description: This use case describes how parking owner view report
Trigger: The user selects the ‘View report’ option in the system
The user must be logged into their account.
Postcondition: The report is successfully display
1 The parking owner access view report
2 System display report revenue, availability of parking space
Activity Diagram
Check in
Figure 3.3: Activity Diagram: Check in Process
The activity diagram shows a Check in process that involves five participants included Customer, System, Parking owner, Barrier, Digital screen.
– Customer:The person driving vehicle pass through the entrance gate of the parking zone. – Parking owner:The person whose parking zone have a new arrived customer.
– System:The system that handle the check in process.
– Barrier:The device used to block the path to prevent vehicles from passing through. – Digital screen:The device that display the check in status located beside the barrier.
1 Driver arrives the entrance gate.
2 System extracts the plate number and checks if there is any reservation.
– If there is a reservation, the system updates the availability of the parking space by decre- ment 1.
– If there is no reservation yet, the system continues checking if there is any account as- signed to this plate number.
• if there is a existing account, system send a check in request to the app of the user and then re-check the reservation.
• if there is not any account, the system will send the signal to the barrier to deny the arrival and show the rejection through the digital board.
3 After updating the availability of the parking space, the system will sends a check in confirmation to both customer and parking owner as well as allows the arrival by lifting the barrier and prompting success through the screen.
Check out
Figure 3.4: Activity Diagram: Check out Process
The diagram shows a Check out process that involves five participants included Customer, System, Parking owner, Barrier, Digital screen
– Customer:The person driving vehicle pass through the exit gate of the parking zone. – Parking owner:The person whose parking zone have a new departure customer.
– System:The system that handle the check out process.
– Barrier:The device used to block the path to prevent vehicles from passing through. – Digital screen:The device that display the check out status located beside the barrier.
1 Driver arrives the exit gate.
2 System extracts the plate number and checks if the parking fee is paid or not.
– If not, the system sends a payment request through the app for user then re-checks the payment status until completed.
– If the payment is completed, the system will update the parking space availability by increment 1.
3 The system sends the signals to allow the vehicles exit the path by lifting the barrier and displaying Success check out in the screen, then sends check out notifications to
Booking Process
Figure 3.5: Activity Diagram: Book a parking space
The diagram shows how the booking process proceeds The participants involved in this process are the customer, system, and parking owner.
– Customer: The person who search and choose an appropriate parking space to make a reservation.
– Parking owner:The owner of the chosen parking space that can have a overview of the booking information.
– System:The component to process the booking process.
1 The customer search his/her destination and their estimated arrival time (optional).
2 The system surveys and lists the possible nearest parking space and then waits for current availability checking.
– If the waiting time is greater than 10 minutes, the system will finish the seeking pro- cess and prompt to user that there is no current available parking space.
– If under 10 minutes, there are any available parking space, the system will provide the recommendation parking space list to the customer.
3 Customer will choose the appropriate parking space and the payment method.
4 After choosing the method to pay fee, the system will check if the method is pre-paid or not.
– If the user choose to pay instantly, the system proceeds to the payment gateway. – If not, the system set payment as pending until the check out.
5 The system updates the payment status and creates new reservation then decreases the parking availability by 1 and sends booking notification to both customer and parking owner.
Payment Gateway
Figure 3.6: Activity Diagram: Casual Payment Gateway
Figure 3.7: Activity Diagram: ETC Payment Gateway
The diagram illustrates the process of paying parking fee The participants involved in this process are the customer, system, bank service and parking owner.
– Customer:The person who perform the payment operation by many given methods.
– Parking owner: The owner of parking space that received payment report and parking fee.
– System:The component to process the payment process.
– Bank service:The service that support the authentication of bank information and trans- action verification.
1 The customer clicks to the Pay button.
3 The customer choose the payment method.
4 If customer did not choose to pre-paid, the system will set the status as pending until the check out process.
5 If the customer chose others method such as banking services or e-wallet, the system will prompt customer to enter the bank account information.
6 The system checks the input is accurate or not.
– If not, forcing user to enter information a gain.
– If accurate, the system saves the payment data and transfer to the bank service to make authentication and verification.
7 The bank service verify the transaction.
– If valid, the transaction is successful.
– If not, the customer have to re-enter the information.
8 After receiving the confirmation from bank service, the system will update the pay- ment status.
9 Customer receives the success payment notification in the mobile app.
Bank service receives transaction commission and sends withdraw notification to cus- tomer through digital banking app.
Parking owner takes the transaction report with parking fee.
1 The vehicle arrives the exit gate.
2 The system checks plate number and queries to database
3 The system calculates the total parking fee, then defines the remain by subtract total to pre-paid fee
4 Check the remaining cost if it equals to zero or not.
– If equal, auto check out.
– If not, request payment to mobile app and operate as the casual payment process.
This section analyzes the overall system, then designs and visualizes an architecture optimized for smart functionalities, and outlines the tools and technologies employed in our smart parking solutions for both closed and open-space parking services.