IOT

Machine Learning-Identifying anomalies with useful insights into operations of the Smart Restroom

M A C H I N E
L E A R N I N G &
D E L I V E R I N G
U S E F U L
I N S I G H T S

T E C H N O L O G Y

ETL – Ideata
Data warehouse – Azure
SQL
Reporting – Power BI

O V E R V I E W
An Internet of Things (IOT) gateway is deployed to capture data from the
Smart Restroom system. High-resolution data is sent to the cloud platform
through secure connections over Bell network. In most cases data resolution
is configurable as per customer requirements, many times frequency between
1 and 15 minutes is implemented.
Challenge was to identify anomalies and variation from thresholds and
provide the operators with timely notifications and useful insight into the
operations of the Smart Restroom.

 

A P P R O A C H
An extensive range of analytics are deployed to convert data into actionable
insight into restroom optimal efficiency and comfort level.

1. Smart Senser will capture the number of people has been using the
restroom through People counter. When the Counter reaches a predefined
number it will generate an alert requesting Maintenance.
– Alert can be distributed via email
– Daily Usage pattern can be reported

2. As an alternative to People Counter sensor, wireless Door Open/ Closed
sensor can be used wherever the restroom is equipped with a main door.

3. It will capture Maintenance visits using the RFID
– RFID swipe In/Out can be reported daily/weekly identifying each
maintenance visit and time spent
– Each Maintenance response time can be reported

4. Sensor Data
– Temperature/Humidity Log
– Temperature/Humidity threshold based alert emailed directly
– Water detection Alert emailed directly- similar to maintenance alerts

5. Happiness Index
– It will capture the feedback through a web GUI on the Tablet
– Daily /weekly happiness index can be reported.
– Happiness Application

 

O U T C O M E
Notifications and Alerts, in the form of SMS and emails, will be
communicated to designated phone numbers/email addresses in the event of
fault/threshold deviations.
A series of widgets are also deployed to provide near real time insights.
The data analysis across portfolio can further help customer make informed
financial decisions about capital investments.