This is the solution to shield retail/hospital/developers’ employees and their customers from coronavirus infection by colorful and meaningful notifications and indirect suggestions to make people decide their decisions positively to reduce the risks.
User ID is user@fake.email, so is password password.
In this COVID-19 crisis, many businesses lost both their employees and customers since both would be exposed to coronavirus infection risk. To brink them back to companies, stores have been doing for disinfection and sanitization, like keeping social distance and putting alcohol hand sanitizers. However, it's insufficient because INVISIBLE infection risk makes people afraid. Both employees and customers want technology to reduce their anxiety.
The best way to defeat fears is to visualize the risk. Fortunately, we live in the world, surrounded by many devices and software services. Cameras and sensors can detect human behaviors and the facility's status, such as recognizing if people flock at the specific area, monitoring how many people wash their hands correctly, and verifying if the staff clean the garbage bin regularly. The cloud providers also give developers elastic and resilient infrastructure services on which developers can build new valuable applications quickly to aggregate sensor data from the edge and visualize them to the employees and customers. IBM Cloud gives us the power to construct worthy applications immediately. It has FaaS, Function as a Service, named Functions on which only things for developers is to write the business logic as a function. Functions can associate with other services, like the Cloudant database and App ID authentication service, without taking the time. In addition to that, IBM Cloud has many open-source-based services, like the Event Streams, that can easily aggregate sensor data.
Here is the thing. Existing technologies help the developers to build applications, but it doesn't ease staff/customers' concerns, which is fear of INVISIBLE infection risk. This issue is what we should solve.
COVSAFE, our application, generates visual-awareness among staff/customers, giving them peace of mind and a safer business environment. COVSAFE makes a risk score or ANSHIN-index, a Japanese word for safety, for each employee and each area of business environment/activity. When ANSHIN-index exhibits a high COVID-risk scenario, COVSAFE notifies the owners/staff in a clear and concise visual notification-format. It helps the team stay clean and maintain a healthy work-life, it reinforces owners to business environment sanitized and disinfected regularly, keeping the customer informed.
COVSAFE concept is "being healthy and clean, keep sanitizing and disinfecting alongside maintaining the social-distance." COVSAFE targets staff+people+operations together, making it unique. The differentiating points are:
- Domain-based or Field-work-based risk calculation
- Personalized/Area-specific risk visualization
- Positive reinforcement visualization to reduce the risks
This is the COVSAFE system architecture.
Here is the flow how the system works.
- Sensors placed at facilities, like supermarkets, send raw data to the edge server. Sensors monitor congestion in the specific areas, people's behavior if they wash their hands, and garbage bins' status if it's disinfected.
- The edge server aggregates and cleanses sensor data, extract parts of data needed, and publishes them to the backend.
- The Event Streams service gets sensor data.
- The Data Recorder, running on the Node.js Functions, subscribes data sent by sensors and pushes them to Cloudant database.
- The Cloudant stores raw data.
- The Risk Calculator, running on the Python Functions, reads data from the Cloudant and calculates the risks of each cell of areas, employees, and garbage bins. Then, it pushes the result into the Cloudant database.
- The Risk Calculator pushes the calculation result to the Push Notification service if the result shows risk high. The push notification service pushes it to mobiles registered beforehand.
- The mobile phone gets notification, kicks our Indicator App. The app calls IFTTT webhook to send the notification.
- The IFTTT service calls the MESH App, running on the mobile, back. The MESH app sends the request to MESH via Bluetooth to change the color. The reason using IFTTT comes from the MESH specification. The only way to turn the LED light on is to kick IFTTT which calls the MESH app.
- The MESH gets the requests from the MESH app. The request includes what color should be changed. The device follows the request, switching the color to tell the current risk that the employees are exposed.
- Employees can see the color LED of MESH device to get known how much SAFETY they are easily.
- Either facility/shop managers or their customers can access our web portal to see how much safety the facilities are.
- The web browser, running on either the computers, the smart phones, or the digital signages, sends the requests to get our COVSAFE web portal.
- The API Gateway passes through above requests if the requests passes the client id authentication.
- The React App, running on Express on Functions, returns components of the web portal, like html, css, and javascript files if the requests pass the user authentication.
- The App ID verifies the user token given in the request from the React App.
- The brower sends requests to not only the React app but also the API service to get data, like the risk calculation result. The API service reads data from the Cloudant database and sends it back to the client.
- As well as the API service reads data from Cloudant, it obtains data, like static files, from the Cloud Object Storage.
Finally, the users, such as the managers and customers, can be informed how mush safety they are and the facilities are via both our web portal and colorful indications.
We already built MVP, Minimum Value Product, and will have a PoC, Proof of Concept, in August with one of developers (the property management companies). On that PoC, we would like to verify how much the COVSAFE works. In deital, we will estimate how the COVSAFE Ads system and notifications eases area congestions. Around this September, we will support multi-tenants by running the COVSAFE as SaaS. Until the end of this fiscal year, we aim to have a PAID Poc with partners. We have a plan to release Beta service till 2021/3 which supports the enhanced monitoring that might be difficult and complicated to detect, like cough and sneeze detections. V1.0 will be released until 2022/3 and launched as a service with a new color-based notification device. We aim it to be smartwatch band that surely be attractive most of people!
Please see Longer version
Please see delivery and try the COVSAFE.
- IBM Cloud Event Streams: used as the Message Bus
- IBM Cloud Functions: used as FaaS on which our business logics run
- IBM Cloud Cloudant: used as the NoSQL database
- IBM Cloud Push Notifications: used to push notification to mobile
- IBM Cloud Object Storage: used as the object storage
- IBM Cloud App ID: used as authentication service
- IBM Cloud API Gateway: used as an API gateway
- Hiroshi Nakagoe - Hitachi Ltd.
- Kentarou Watanabe - Hitachi Ltd.
- Santosh Maurya - Hitachi Ltd.
- Shin Tezuka - Hitachi Ltd.
This project is under the Apache 2 License. See the LICENSE file for details.