Categories
Open YPEC 2021

ON01-Tryonn

Tryonn is a SaaS company. We help fashion businesses to develop AR solutions for consumers to try their products online. Tryonn’s driving technology is our AR try-on technology to help users try on fashion products. Consumers are expecting sophisticated and engaging AR-shopping experiences, but the AR-commerce solutions in the market now are not corporating comprehensively in AR-shopping for development and marketing, which AR can only be used as an additional feature. Our AR try-on technology is comprehensive in that it supports a wide range of fashion products and textures, and the AR effects can be used as marketing tools for fashion sellers in social media to promote their products. The charging mechanism of Tryonn is based on the sales commission rate; The commission rate at the current stage is 10%.

Categories
Open YPEC 2021

ON02-Optimizing Station Design through a Crowd Simulation Model – MassMotion

Human factor is a core consideration in transport planning so as to safeguard public safety. At the early design stage of a mega project, interactions between human behavior and future provisions are complicated to be understood, taking railway stations as an example. In order to demonstrate smooth passenger flow, acceptable Level of Service (LOS) and evacuation time to be within required design limitations during traffic peak hours at a critical design year, passenger flow simulation shall be conducted to verify the practicality of a station design. Human behavior can then be better understood using a crowd simulation model – MassMotion. MassMotion is an in-house pedestrian flow modelling software developed by Arup. Its result output can be in formats of screenshots, maps and 3D animation. These output provide realistic visual experience for our clients, designers, architects and engineers to identify critical areas for design improvement, such as bottleneck, congestion or conflicting movement from a passenger viewpoint walking inside a railway station. MassMotion has been widely used in various projects around the globe as an engineering solution to enhance design performance. In view of the COVID-19 pandemic, usage of MassMotion has surged to simulate social distancing situations using its Proximity Analysis tool.

Categories
Open YPEC 2021

ON09-Artificial Intelligence (AI) Based Accident Prevention System for Escalators

In past 2 years, an average of 70% passenger accidents in MTR network were related to escalators, while about 60% of them were related to passenger behaviours including losing balance, carrying bulky objects and influence by alcohol. “AI-based accident prevention system for escalator” aims to identify high-risk passengers, predict passengers’ behaviour and trigger alert to safeguard passengers using escalators in MTR stations. The system utilizes Light Detection and Ranging (LiDAR) sensors to detect speed and volume profiling of passengers under crowd with privacy guarantee. The AI algorithm identifies characteristics of passengers and predicts path when high-risk passengers intending to use escalators instead of lifts after passing entrance gates. The system will initiate relevant public address announcement together with visual alert message to remind these passengers to use lifts. Instant message will be sent to station operators to provide timely assistance. This system would also educate public to adopt safer behaviour when using escalators. The big data collected would also introduce value-added feature on passengers’ flow pattern and behaviour for crowd control. System functional test has been completed in Choi Hung Station (Pilot Site) with satisfactory results. System fine tuning is in progress. Trial operation would be conducted in November 2021.

Categories
Open YPEC 2021

ON08-Transformation of Smart Toilet through Internet of Things (IoT)

The invention of toilet facilities can be traced back to thousands of years ago in our history. Development of toilet facilities represents the evolution of mankind civilization, rather than adhering to the norm, the Architectural Services Department – the Government of the HKSAR is evolving from traditional toilet into ‘Smart Toilet’ with the application of the emerging Internet of Things (IoT) technology. IoT connects interrelated devices through network to collect and transfer data in conglomeration without human involvement. In Smart Toilet, various types of sensors are employed to retrieve useful data for facilities upkeeper to be aware of and monitor the conditions of toilets efficiently. For example, the air quality sensor allows real-time monitoring of indoor air quality which facilitates the cleaners to response and upkeep the toilet’s hygienic condition. With the help of Artificial Intelligence (AI) and Big Data analytics, the massive data retrieved from IoT can also provide greater insight to guide the enhancement of Smart Toilet such as better user experience and lower energy consumption. The success of Smart Toilet acts as an important threshold for the future development of “Smart Building” and even “Smart City”, which transforms the world into a better place.

Categories
Open YPEC 2021

ON07-Automatic Radar-based Intelligent Shunting Application (ARISA)

The use of smart sensors has been a game changer in modern railway business. Through digitalization, railway performances become measurable and observable in ways never seen before. Human factors during train operation can also be quantified and improved, minimizing the number of unwanted incidents and injuries due to human error. An automatic and standalone device has been developed by MTR engineers to alert train captains during train shunting. The design employs a high caliber microprocessor, an integrated RADAR sensing module, a SD card module for data recording, and self-implemented data filtering software for smooth detection and quick response. The solution is also highly agile in various environment, including tunnel, outdoor and depot areas. Compared with the conventional shunting operation that solely depends on manual operation, ARSDA further allows the data analysis of train captains’ performances using records of each shunting operation. By plotting the data against a safe driving profile, the operation performances can be evaluated based on the recorded approaching speed and acceleration. With multiple tests already conducted, these data are currently in trial as a part of training for train captains, as safety is always the ultimate goal of daily operation in MTR.

Categories
Open YPEC 2021

ON06-Pathogen Prediction Modeling

Arup has collaborated with the City University of Hong Kong to develop a pathogen prediction platform. The platform adopted the AI, IoT and cloud technologies, which ease the deployment of the IAQ monitoring platform into new and existing buildings. The team had successfully overcome the technical barriers in the industry and made pathogens ‘measurable’ and ‘predictable’ at building level with a financially and technically feasible approach. The team actualizes the knowledge regarding pathogens and transmission within built environment and transforms it into tool for prediction and quantification of infection risk with an affordable and technically viable solution. The AI algorithm were supported with pathogen correlation research with onsite sampling, and it has successfully identified the correlations between potential pathogens and commonly measurable parameters in air. Apart from the correlation, the algorithm could also predict the air pollution and thermal comfort trend up to 30 mins with an accuracy of 90%. The HVAC and the purification system can then have sufficient respond time to adopt appropriate adjustment to ensure the stable and excellent air quality. From project experience, the platform helps to reduce more than 1/3 high risk hours during operation period.

Categories
Open YPEC 2021

ON05-Smart Air Quality (SAQ) Monitoring System

The Smart Air Quality (SAQ) Monitoring System is a breakthrough solution designed for efficient environmental monitoring in the railway industry. It operates with various types of LoRa sensors measuring multiples parameters, including temperate, humidity, CO2 and NO2 with the customization alarm of measured parameters. This system adopts the plug-and-go operation of sensors, in which registered sensors connect to the LoRa network whenever they are on. Thus, it provides a high level of flexibility and scalability of installation in brownfield with the benefits of LoRa wireless technology – long-range and low power.

For efficient analysis of collected data, a visualization tool, including a dashboard, heatmap, and indicators, is created for users to remotely view the real-time data. Based on the data collected, correlations between the air quality data and surrounding variables can be identified and insights can be generated, e.g. platform temperature and patronage. The more diverse data collected, the more comprehensive analysis can be achieved. Therefore, responsible parties can ride on the identified trends and work on the follow-up action(s).

Overall speaking, we can obtain more comprehensive environmental data, raise staff and customers’ awareness of air quality, and ultimately enhance the air quality condition and public health in MTR premises.

Categories
Open YPEC 2021

ON04-Medium Access Control with Efficient Medium Utilization

Internet of Things (IoT) enables the possibility of the smart city in which devices and environmental status could be remotely monitored, managed, and controlled. Low Power Wide Area Network (LPWAN) connecting a large number of devices in a large area with high energy efficiency is the cornerstone technology in an IoT system. Several potential IoT solutions have been proposed to meet the requirements of LPWAN, including LTE-M and Bluetooth Low Energy (BLE). LoRa has the features of large coverage, low power consumption, low cost, and resistance to noise. These physical properties make LoRa be the next emerging IoT solution for building LPWAN. However, the existing LoRaWAN medium access control (MAC) only offers the pure ALOHA method designed for low duty cycle applications. Severe collisions are expected if a large-scale system is launched. Therefore, in this project, multiple MACs with the features of pure ALOHA, slotted ALOHA, IEEE 802.11, IEEE 802.15.4, and channel activity detection (CAD) were proposed and tested by experiments. The performances of the MACs were evaluated by comparing the packet delivery ratio (PDR) and data throughput in bits per second (bps). Almost 100% PDR and a maximum of 4800 bps were achieved by one of the proposed MAC.

Categories
Open YPEC 2021

ON03-Application of the Concept of Digital Twin in Enhanced Design of Thermal Management in Existing Data Centres

Data centres annually consumed over 4,000 TJ in Hong Kong, in which over 35% attributed to air-conditioning systems. Computer Room Air-conditioning (CRAC) Unit is normally adopted for data centres. This conventional passive air conditioning design tends to be segregated from external environment, such that indoor air environment could be controlled rigorously with a view to achieving required resiliency and reliability. Hence, hot air from server racks would require to be treated instead of extracted directly outdoor, and utilization of free cooling is often not considered. This inefficient thermal management is usually overlooked because it is always a challenge when analyzing the complexity of air diffusion and temperature profile of data hall with the stringent uptime requirements. Concept of digital twin is introduced to overcome the problems, which is an innovative approach to precisely and effectively modelling data centres’ operation behaviour integrated with real-time data collected by IoT sensors and Computational Fluid Dynamics (CFD) simulations to enable detailed thermal analysis in “what if” scenarios with zero risk to data centre’s operation to accomplish an optimal energy saving and carbon reduction solution with 30% anticipated electricity saving, and this approach can be applied to all existing and new data centres.

Categories
Open YPEC 2021

ON10-Smart Watchdog for Human Factors in Railway Systems

Human errors are undoubtedly the pain points in railway operation, raising concerns about safety and services. As one of the world’s leading railway operators, MTR is always exploring innovative opportunities that help improve the safety and services. Using system data to monitor operators’ action can prevent these incidents, however interfacing with mega systems like signalling and power control systems requires expensive modification. The latest video capturing technology enables cost-effective collection of data directly from the screens of operators without affecting the system operations. It also facilitates the design of a generic interface for those mega systems. Additionally, video analytic techniques like deep learning and text recognition are performed on the video streams to alert wrong actions from operators. We collectively name this type of application as Smart Watchdog for Human Factors, with two examples as follows. 1.Railway Route Setting Monitoring System interfaces with signalling system to provide instant warning when an intention to violate the predesigned route setting rules is detected by deep learning. 2.Power Switching Communication Enhancement System interfaces with power control system to enhance the communication during power switching operations and alert operators when wrong power switching execution is detected by text recognition.