Foreword
Opening the business series of blog publications focused on real-time parking management, we thought it would be best to begin by outlining quality and up-to-date perspectives of industry professionals, rather than describing our own view of the space at length.
The smart parking space is very broad and information sometimes seems fragmented, so the plan was to write an overview of few publications capturing salient points. While combing through materials on the web, we came across an extremely comprehensive article in ITS International – encompassing lessons from history, technology alternatives available on the market and achievable and pragmatic business goals for smart parking.
We shortened the original The Smart in the Smart Parking* article for the purpose of our blog, keeping all findings and conclusions but omitting some secondary details.
The title however was changed to reflect Nwave’s view on the subject. More about that in our short commentary at the end.
Nwave’s Summary of ITS International’s THE SMART IN THE SMART PARKING
There are a lot of technological solutions to alleviate the urban traffic crunch and make on-street parking more efficient and better managed. Cameras and sensors are the two most common alternatives. This article discusses the advantages and challenges of both.
In North America, communities are taking up new vehicle detection technologies to curb the parking crisis. It is a notable part of broader Smart City initiatives to change the transport culture within the country. This is achieved through ITS solutions and policy changes which will reform parking by enabling those without any other option and discouraging most people. The potential for success has been proven in some cases, while in some other cases, it has not. Some cities like Washington, DC, Los Angeles and San Francisco have displayed variable success rates with the use of these smart parking systems. The differences in success rates are attributable to the various technologies for detecting parking spot availability, which include modern cameras and sensors.
Detecting parking space occupancy
According to a transport analyst, Mark Zannoni, smart parking system must have the following functions:
- Identification of free parking spots and providing that information to users on the road
- Multiple options of payment for parking services, which include mobile payment
- Improved enforcement methods to ensure there are no significantly overstaying vehicles
- Relay data to the transport authority of the city to be used for policymaking
Four key functions of a smart parking solution. (c) Nwave
Real-time parking space availability monitoring is most commonly solved with sensors and cameras. Sensors can be inserted into the road and utilize radar or magnetometers to detect vehicles while overhead cameras installed on the light poles and rely on computer vision software for vehicle detection and pinpoint open spaces. Vans and trucks can be equipped with mobile parking detectors. The data collected from these systems can be analyzed to produce useful statistics, forecasts and insights that will be valuable to local authorities when it comes to policymaking.
Vehicle detection sensors vs. Fixed over-head cameras
Wireless sensors are inserted in the ground, under the pavements where they will not interfere with line of sight and will not be affected by weather changes. They work by utilizing vehicle metal detection in parking spaces using magnetometers, but more precise sensors that use radar detection are also currently available. Wireless sensors have a low cost of maintenance since the battery life is long, thus does not require regular changing, and the product has a clear single-aim engineering design. Despite all these pros, vehicle detection sensors have some limitations because they are in-ground. They do not provide any other data other than space occupancy, therefore, they are a single-goal technology that have lower returns on investment. Replacement or maintenance of a sensor will also require it to be dug up from the pavement.
Fixed cameras have the advantage of monitoring several parking spots at once, usually a street lane or a significant area of a parking structure. They are also instrumental in enforcement of parking rules and policies through the ability to detect license plates and capture violations. The most significant advantage of cameras is that they can be used in management of traffic as well as surveillance.
However, the expense of purchasing and maintaining cameras is significantly higher than that of in-ground sensors. They need power and internet to work and are affected by vibrations, weather and are prone to vandalism. The multipurpose usage also has notable limitations due to various focus, viewing angles and resolution requirements of the different applications. Being much more nuanced than sensor technology, pervasive video analytics comes up as a more difficult buying decision. However, choosing the right solution depends on the needs of the city and the main problem that needs to be solved.
Lessons learnt from successful implementation in Washington
Early implementation of smart parking management sensor system in San Francisco and Los Angeles was unsuccessful due to the issues with detection accuracy, scaling and reliability. The program was, however, successful in Washington. The assistant director of transportation operations and safety for District Department of Transportation (DDOT), Soumya Dey, thinks that time and progress in engineering are the main factors that led to this success. The technology for developing sensors has improved and better cameras are available leading to a decline in prices since the LA deployment.
DDOT analyzed the problems that faced the former large scale smart parking roll- outs and realized that cost of on-street vehicle detection was the major drawback, which meant that a cheaper way of implementation would improve chances of success. Smart parking management in Washington flourished because one sensor was used for every two parking spaces, unlike the other cities where each space had its own sensor. This was a cheaper option that still provided a lot of valuable data. DDOT also abandoned the variable real-time valuing and opted for monthly or quarterly pricing adjustments.
The aim of the DDOT, according to Dey, was to ensure that there was always one free parking spot at all times per city block. Their priority was to reduce congestion in a city where traffic was caused by drivers looking for a parking space 30-35% of the time. The DDOT then initiated a six-month trial program in various areas to gauge the reaction of drivers to the new pricing structure. The pilot variable pricing scheme began in 2016 where 1,000 spots were fitted with 500 vehicle detection sensors. The program provided information on how drivers would respond to the pricing changes and the effect of their response to congestion.
Settling on the correct choice
DDOT utilized various technologies in different phases of their pilot program in Washington, DC. Mobile cameras were the best in terms of providing a foundation to be used in determining need for a parking space and behavior of drivers. Sensors were then permanently fixed under pavements and cameras mounted on light poles to ensure that the sensor network was comprehensive. According to Dey, these precautions were taken to prevent reoccurrence of past mistakes and to address the specific needs of the city.
It is very pertinent for municipalities that want to adopt smart parking management to re-evaluate their main goals: whether it is reducing occupancy, decongesting, increasing revenue collection, and cutting parking enforcement staff or some other objectives. Smart parking is not just one “plug and play” solution that universally fits all and with continued technological evolution of vehicle detection, it is becoming increasingly difficult to choose.
Unpromising pilot program in California
In 2011, LA initiated a smart parking management system, Express Park, which involved 7,000 spaces on the street with the aim of varying pricing and directing drivers to open spots. This magnetometer-based smart parking sensor program did not achieve its goal.
Dependability and Precision
Deployment of one sensor per spot was too expensive and the project could not scale past an initial 400-space pilot area. There was interference between the sensors and the accuracy of the program did not live up to its predicted value of 95%, leading to public distrust. In San Francisco, the installed sensors had interference from the light rail and bus systems, resulting in dependability and precision issues. Despite efforts to fix these problems, the programs have not yet recovered.
Nwave Commentary:
In conclusion, it is important to understand three main points critical for any smart parking solution: accuracy, reliability and cost. These three points are indicated to be the main impeding factors for wide technology adoption. Of course, the cost vs. accuracy dilemma makes a buying choice extremely difficult. The value of the higher detection accuracy or downside of lower cost compromise with being “half accurate” are not always easy to explain in business metrics and ROI terms.
Nwave’s answer to this problem is simple. A “no-compromise” smart parking solution bringing both the highest accuracy with unparalleled low initial investment and fast ROI. It’s a no-brainer how this is ideal as a business solution. Plug and play smart parking has just arrived!
Stay tuned to our blog to learn more!