Data Flow through Google Services
Device sending updates to Google IoT Core
Benefits of Preventive Maintenance
Pakistan has a vast array of industrial and manufacturing setups churning the economy at various level but they lack quality systems for monitoring, optimization and fault detection which result in costly downtime for them. These setups are unable to afford multimillion dollar solutions by industry leaders. When approached by a particular setup for evaluation of the process for optimization we realized that they can utilize a system that is cost effective and not only helps the plant reduce downtime but increase efficiency as well by observing and reporting back on various legs of the process. We realized the market is wide open for a solution which solved not only monitoring problem but let the operators know of an impending failure before it happens!
What it does
The system monitors the process of a manufacturing plant through industrial grade sensors. The data is relayed to a Cloud IOT platform, in this case Google IoT Cloud where it is converted into a visual presentation for the operators. The system also provides a capability of preemptive fault detection to operators via web app, sms and mobile apps. This enables the operator and owner to get ahead of the problem and avoid valuable downtime resulting in better profitability.
How it's built
Industry standard rated sensors are deployed at a plant which gather data and relay to a central unit. The central unit relays them to cloud via redundant paths. Primary is WiFi, the secondary is GSM module able to relay data down to 2.5G speeds and tertiary is onboard data storage with less frequent updates sent via SMS text message. The cloud platform stores and processes for visual representation to monitoring teams. The same data is also parsed through machine learning & data analytics algorithms which evaluate and monitor the process constantly for preventive maintenance updates.
Challenges we ran into
Cloud Integration while they have been made very easy yet the process is not as simple with multiple way of accomplishing the same task, its a challenge to find a method that suits. Selection of IOT Cloud was also a factor.
Accomplishments that we are proud of
We have ready hardware which has been field deployed in actual industrial units via bluetooth and data logging channels. During the course of the JAZZ SSDG Hackathon we updated the code to be able to connect to Google IoT Cloud and be able to send basic data which can be stored and consumed via cloud database and data representation tools.
What we learned
IOT Cloud integration. Cloud Data Storage, representation and data analysis tools. Data Modeling strategies for incoming data and how to sell the features as separate services on demand.
What's next for IOTO
IOTO is evolving into an online realtime data acquisition system with capabilities to send and store and data to IoT Cloud. Next step is to train predictive models on realtime data to be able to send predictive maintenance alerts to operators. The models would evolve periodically and would require quarterly updates.