Wildfire Prevention Tool

This Tool aims to pre-empt chances of Wildfire and mitigate the hazard using Pega CDH and case management

Comment 1

Inspiration

Wildfire is a common problem which has risen due to Global warming and climatic adversities in recent years. We randomly see occurrence of wildfires in different global regions causing the destruction of flora and fauna and forcing governments in different geographic terrains to lose millions. As per internet research , From 2012 to 2021, there were an average of 61,289 wildfires annually and an average of 7.4 million acres impacted annually. In 2021, 58,968 wildfires burned 7.1 million acres. As of September 2, 2022, nearly 48,500 wildfires have impacted about 6.2 million acres this year. To control this worldwide acute problem , the aid of Digital Technology is taken. A Digital Tool , Wildfire Prediction Tool is an Application built out of Pega platform and Pega Decision Hub which will come handy to Predict the occurrence of a wildfire and alert government bodies to take requisite pre-emptive action to mitigate this calamity . UNSDG Goal 13: Climate Change & Goal 15: Forest and Biodiversity are the 2 main pillars which serves as an inspiration to this tool.

What it does

The Wildfire Predictive Tool is a futuristic application composure of IOT devices and Pega CDH.

Trees in the Official Wildlife zones would be fitted with External sensors which will initially read the climatic condition reading and transmit the data wirelessly to the Weather Forecast centre. The Ground Temperature and Relative humidity records are recorded by the sensors and transmitted wirelessly to the data centre. Once the data is fed into a pega AI engine, it predicts the probability of the forest fire. Once the probability crosses a configurable threshold concerned departments are alerted , so that they can take preventive measures.

How we built it

To implement this solution , primarily trees in the Official Wildlife zones would be fitted with External sensors which will initially read the climatic condition reading and transmit the data wirelessly to the Weather Forecast centre into the following way:

1.Data is received from DHT22 sensor by the arduino uno 2.This data is sent to the arduino uno at receiving side using the NRF24 module 3.The data received at the receiver end is transmitted to the Lolin NodeMCU via serial communication port 4.The Lolin Board is connected to the wi-fi. It pushes the data to the Pega Server through a rest API in an interval of 15 seconds.

This Data will be used by a Pega Enabled Decisioning Application which should have a Predictive model in place. This Predictive model will be fed with previous data of Wildfires for past 30-40 years with all relevant climatic data. Based upon the input, the model will try to predict a percentage of wildfire.

Once this prediction score crosses 45 ,there should be an Automated Case creation of a Parent level case. Once Case creation is done , there will be parallel sub-cases created and assigned to different work-queues based in different Government bodies like: Fire Station, Police , Disaster Management

Now based upon this case data ,different bodies should enact and proceed the sub-cases stage-wise towards resolution.

Once these sub-cases are resolved, the Parent case will be closed with necessary pre-emptive actions to mitigate the Wildfire.

 

Challenges we ran into

  1. Primary challenge was lack of real-time infrastructure and integration channels . To mitigate that we had to depend upon simulations
  2. For the sake of demo , we had to use stubbed data , as real time data gathering was not possible to obtain.

Accomplishments that we're proud of

  1. We were able to think through an idea ,implementing which we can develop a system which should be capable of saving lives. This idea ,if implemented properly will help in restoration of Climate , save lives and save economic disaster.

  2. We were able to use the features of Pega CDH and Case Management hand in hand.

  3. All the components used were fully OOTB without any custom coding done.

What we learned

  1. We studied various aspects of Pega Decision features and were able to use proper predictive model for use.

  2. We have compared different sensor types available in the market and were able to finalize DHT 22 as the intended model.

What's next for Wildfire Detection Tool

We will work this idea to give a proper MVP shape and try to incorporate add-on functionalities like :

  1. Prediction of path of a wildfire. 2.Omni-channel communication with different government bodies for notification of Wildfire.

Built With

+ 2 more
Share this project:

Updates