Database and IT project managment

IoT Smart Parking Meters

Project Proposal

Introduction

Westminster, Maryland is a quaint college town located northwest of Baltimore, Maryland and just south of Gettysburg, Pennsylvania. Main Street features various attractions including many unique, trendy restaurants, small boutique shops, and the local library. At any given time of the day, Main Street is packed with people and their vehicles. These people keep the local economy booming with their purchases, but they also help the town of Westminster by paying to park. It is rare to find an open spot to park on Main Street due to the sheer popularity of the area. The town has the ability to use the rarity of parking places to bring in additional revenues.

If you drive down Main Street in many cities and towns in America, one apparent feature is the parking meters. These coin-operated meters have been used for decades and very few changes have been made over time. Like many small towns in America, Westminster still utilizes outdated, manual, coin operated parking meters. These meters must be tended by a parking enforcement officer. In our increasingly technical world, these meters are neither user friendly nor cost effective.

Although the technology of these meters is outdated, parking meters themselves can still bring in a lot of revenue for a city or town. According to Krohe (2012), Washington, D.C. typically makes $40,000,000 a year in revenues from parking meters and parking meter related fines. With profits of this magnitude, it is difficult to disagree that parking meters can still be an effective means of revenue. Westminster is definitely much smaller than Washington, D.C., but there is still a great opportunity for revenue by utilizing metered parking effectively.

Unfortunately for the town of Westminster, revenues are not currently maximized with the current meters. There are too many meters to be managed by the few parking enforcement officers that the town has employed. As a result many people are able to park for free and leave before a fine can be assessed. The parking meters on Main Street have the potential to bring in large revenues for the town of Westminster. In order to maximize profits, changes will need to be made and new technology will need to be integrated.

Additional parking enforcement officers could be hired by the town of Westminster, but the increased cost of hiring additional employees would likely negate much of the overall profit gained. The meters themselves must be upgraded in order to increase profits and reduce operational costs.

Statement of Need

In public areas and shopping malls, it is often challenging for visitors to find a vacant parking spot. On working days and during busy hours, visitors must allocate additional time to enter and exit parking areas or risk running late to their appointments. This paper will demonstrate how the IoT connected parking meter, its associated management platform, and parking mobile app will mitigate the issue of overutilized parking while providing the municipality with more revenue and improving the customer's experience.

While coin operated meters are a proven way to manage overutilization of public parking while generating revenue, they have some shortcomings. The first of these is maintenance costs and timely service. At any point in time, a percentage of parking meters are out of order. While a meter is out of order, not only does the municipality lose money from the lack of paying customers, but the lack of a spot can also result in customer frustration as there is often uncertainty regarding the legality of staying parked in a spot with an out of order meter. Conventional coin based systems rely on periodic service inspections and customer reports for inoperable meters in order to generate a service request and address the problem. This solution is inefficient as there is no guarantee a defective meter will be reported let alone repaired in a timely manner. IoT smart meters will address this problem by performing routine diagnostic requests against deployed meters and automatically dispatching a service request to any defective or non-responding meters.

When parking, many customers will overfeed the meter in order to avoid having to come back and top it off later. This results in a frustration for customers who are often paying for more minutes than they strictly need as well as a reduction in revenue for the municipality as a result of customers who free ride off meters filled by the person previously parked in the spot. Smart meters address this problem by allowing customers to pay precisely for the time they use, no more and no less.

Parking lot overcrowding is largely a result of an information asymmetry between those leaving spots and those looking for open spots. In addition to frustration by customers who are unable to find spots quickly, crowded parking lots can be a surprisingly dangerous place as many of those looking for spots may be driving forwards, backwards, or turning spontaneously with little warning. A system that identifies open spots and leads drivers to them through the use of live updates in a mobile application as well as networked digital signage (placed at strategic entry/exit points) would dramatically cut down on parking lot overcrowding while also making them a safer place.

Project & Requirements Definition

The solution to the aforementioned parking issues and inefficiencies is to implement IoT Enabled Smart Parking Meters. The primary functionality of the system refers to:

Detect the presence/absence of a car in the parking lot
Record the parking date and time and the leaving date and time per car as well as the license plate, make and model of the car
Offer drivers a mobile application for two operating systems (Android and iPhone) for making payments
Process payments
Record parking violations and expired parking time
Communicate parking violations to local police department for issuing fines
Keep parking information for data mining and analytics.

The technical requirements have both hardware and software components.

The hardware refers to obtaining approval from the city, purchasing, installing and testing the ultrasonic sensors & master controller for parking places.

Software requirements are complex as they need to have the following functions:

Receive and process information from the parking sensors
Receive and process payment information from mobile devices (smart phones, tablets, smartwatches, and smartglasses)
Calculate parking time and compare to parking payment
In case of parking violations, transfer information to police station and generate fine
Email to end user the parking ticket-payment, or the parking fine
Generate reports for parking situation (available or not available parking) information that will be transferred to digital signs placed at the parking area entrance
Generate revenue reports (from both park ticket payment and parking fine events) save information in a data warehouse on a daily basis for reporting purposes
Generate reports from the data warehouse (e.g. parking availability per day of week, time of day, season, and holidays)

Implementation Function Proposal

There are three major components for this scenario which are parking management server (PMS), mobile App and IoT platform. PMS is used to monitor all the parking spaces and also manages billing and metering for all users. IoT hardware platform connects all the parking places to PMS. The hardware platform also detects the absence or the presence of a vehicle. The IoT hardware platform is powered by Arduino Yun. It uses ultrasonic sensors so as to detect the absence or presence of a vehicle in a parking space. The PMS component is usually implemented as an application server which runs on Python. The component also keeps track of all the devices and manages reservations and billing. The mobile App which is also known as Auto Park is a natively written android and iOS app. The PMS application server is hosted on the IBM Bluemix cloud platform (IBM, 2016).

After the mobile app is launched for the first time it has to ask for the license registration number of the user’s vehicle. This acts as a unique identifier for the PMS to track the app for billing purposes. The app sends requests to the PMS so as to get the status of all the parking places. This can be used to display the map where each space in the city is identified with a slot number such as 001, 002, and 003.

A user approaching an unoccupied parking space can reserve the parking spot in advance by tapping on the desired vacant parking spot on the mobile app. This in turn prompts the PMS to initiate a billing session for the user. After a reservation is made the user can finally pull in and parks their vehicle in the designated parking space. The hardware then sends the status update to the PMS so as to indicate reservation confirmation. After the user pulls his/her vehicle out of from the parking spot, the accumulation of billable time is stopped, payment processed, and a digital receipt is generated.

The system solves the most frustrating aspects of parking . With the system in play, urban users do not need to search for parking spots. Drivers are also able to observe the current state of parking spot queue as they drive along the township or city. In addition, the system allows the drivers to accumulate information on the current state of the city’s network as they travel. They do not need to transverse the whole network. Information is also exchanged between two drivers concurrently located at a node of a certain network. In addition drivers incorporate new information into their belief about the state of the system. Urban drivers use this information when making decisions on where to park after they reach their destination. In addition, the system can provide real time updates to the citizens and the authorities (Chavan, 2007). Lastly the system can automatically assists in track billing based on the driver's actions as the vehicle pull in or out of parking spaces.

Implementation Technical Proposal

The implementation of our IoT enabled smart parking meter solution necessitates the use of a tightly integrated and failsafe technology stack. At the hardware level, our parking meters will be manufactured to accept a variety of payment methods including major credit cards, NFC based mobile payments, smartphone app based payments, as well as conventional coin based payments (coins will be discouraged through the use of a minor discount for other forms of payment). Meters will have integrated ultrasonic sensors which can be used to detect the presence or absence of a vehicle in the allocated spot. The use of sensors will allow customers to save money by refunding them for prepaid unused meter minutes. Any loss of revenue due to the refunding of unused parking minutes will be offset by the increase in revenue due to customers being unable to capitalize on meters with leftover minutes from the previous customer. A custom low cost circuit board running an ARM processor and loaded with a remotely updateable firmware will be integrated into each parking meter.

Rather than inconvenience local municipalities with breaking concrete in order to install underground wiring and communications infrastructure, our IoT enabled smart meters will utilize existing low voltage wiring infrastructure used with conventional meters. Networked communications will be sent in real time over a mesh network utilizing a combination of a low-rate wireless personal area network ( in accordance with the IEEE 802.15.4 LR-WPAN specification) nodes and pre-existing low voltage power lines. A dedicated network controller will need to be installed for every 250 parking meters or quarter mile of distance from the nearest controller. The controller will aggregate payment, timing, diagnostic, and parking availability data from the nearby meters. This data will be converted into JSON format, encrypted using a 192 bit AES encryption algorithm, and then transmitted over the public internet to the nearest company data center for processing.

A load balanced and scalable infrastructure of application servers will be set up inside each data center. These servers will be listening for inbound data packets containing real time parking, payment, and customer data. Occasional polling will occur of the smart meters using an outbound HTTP request. Meters that fail to respond, will be internally flagged as possibly defective or out of order and a service request will be dispatched to a local contractor for further inspection.

Application servers persist transactional data such as customer records and payment information to a cloud-based relational database for rapid real time retrieval. Analytics and historical information will be stored on a Hadoop NoSQL database. Tableau will be set up as a Business Intelligence (BI) tool to periodically run batch analytics processing jobs against the data stored in the Hadoop database cluster. Tableau will use a variety of Commercial Off The Shelf (COTS) and custom written functionality to generate business insights into data. Metrics on parking will be produced including figures such as average parking vacancies per city block, mean and median parking times, and user correlative data allowing for targeted promotional marketing to those customers most likely to utilize the service.

Amazon Web Services (AWS) Mobile will be leveraged as a service to host a cross platform iOS, Android, and browser based mobile app for users to manage their parking information in real time. An application will be deployed in the AWS cloud allowing for users to query for real time open parking spaces near their current location. Additionally a push notification service will be provided for users of our mobile app allowing for friendly reminders to re-up parking meters when they begin approaching expiration. Payment information for the parking meters can be stored inside the application enabling quick and convenient meter payments without requiring customers to return to their car.

Development Plan Approach

One of the essential aspects of a project is the creation of a solid development plan that encompasses a timeline of critical tasks and milestones. Some example tasks that will be applied to this project are establish project requirements, tasks planned, define mission critical tasks, individual team member duties, assign tasks to team members, work schedules, software requirements, testing, updating, implementation, and project completion date. This is not a final list as there will definitely be more tasks as the project begins to shape itself in the upcoming weeks. The software MS Project will be used to monitor the life cycle of the project and meeting deadlines. There will be a breakdown of expected days to complete each task, the beginning date, end date, and the specific task number. Some tasks can be completed independently, while others will have subtasks that cannot be completed without the prior one being accomplished beforehand. At this time the project can be completed with the following six step plan: planning, development, testing, updating, implementation, and completion. As most of these steps are self-explanatory, the implementation phase involves putting the system into operation while providing guidance and troubleshooting until the completion phase, which consists of the organization taking full control of the system without oversight of the development team. Dealing with a project like this, there is a desire to have milestones to keep the team motivated and the customer satisfied with the progress in an attempt to provide them with some visibility of the project. A few milestone examples would be transitions between phases of project plan, project completion percentages (25, 50, 75, and 100), completion of software, and final implementation of software.

References

Chavan. (2007). Planetizen contemporary debates in urban planning. Washington: DC: Island Press.

IBM. (2016). IoT Enabled Smart Parking Meter. Retrieved February 18th, 2017, from Bluemix blog: https://www.ibm.com/blogs/bluemix/2016/12/iot-enabled-smart-parking-meter/

IoT Enabled Smart Parking Meter - Bluemix Blog. (n.d.). Retrieved from https://www.ibm.com/blogs/bluemix/2016/12/iot-enabled-smart-parking-meter/

Krohe, J. (2012). Hunting for dollars. Planning, 78(8), 26-30. Retrieved from http://ezproxy.umuc.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=f5h&AN=82509953&site=eds-live&scope=site