Situational Awareness is a long known term in the military industry, it’s a key component in the resource allocation and identification of hot-spots.
Phumus primary source of information is a network of wireless environment sensors with ranges up to +20km, these sensors are extremely low cost, maintenance free, autonomous, easily deployed. The sensor network provides environment information in real time.
Phumus secondary source of information is the integrated and tracking module for the response emergency services, the network provides granularity from the control center in charge of an incident, the means involved and finally the individual in the field. These sensors are based on the same characteristics as the environment sensors, low cost, maintenance free, autonomous and small.
It’s also possible to integrate Phumus with 3rd party data provides, emergency services, weather services, civil protection, private automation and other related services.

All the sources of information combined will provide an accurate and true Situational Awareness, a true and real model that is available to the decision maker and those involved in the incident, this combined information is also feed into the AI and machine learning platforms.

Artificial Intelligence and Machine Learning are big words in the recent times, this is not without reason. Machine Learning and Artificial Intelligence are able to provide realistic models of action and reaction, training scenarios and optimal assets response.
The data collected by all the sensors and 3rd party consumers is used to train realistic models, with this models we can predict behavior in high value location, whether located in forest, industry or cities. Emergency services will know where to deploy in the most effective way and will also be able to provide feedback to the current model thus improving the future models.
How do this works?
Imagine a scenario that in a specific location the wind direction usually shifts from the morning to the afternoon, in the past this model needed to be programmed. With the usage of machine learning the pattern can be “learned by the system”, now this is a very basic use case, taking in consideration all the variable in a wild fire is a enormous task due to the sheer number of variables. Phumus is the right solution for this kind of analysis. It works round the clock verifying and maintaining all the variables in check.

Generated models identify zones of high risk, high value and possible response time, with this data and reporting hot spots can have a better response approach. During predicted or announced events the response or mitigation assets can be pre-deployed or on warning state.
Prevention takes the most important role on the chain of events of an incident, being able to identify and predict behavior mitigation can be put in place before the event.

In case of an event, Phumus the event with all the relevant information, location, environmental metrics, impacted sensors and this while there event is rather small. Before the first emergency assets are deployed the possible solutions scenarios are available with all the relevant information, and predicted incident behavior. The decision maker will have all the data at his hands from the first moment enabling the correct decision based on facts.
Being able to see the hot spots, the mitigation agents, and all the environment metrics in a single dashboard enables the resources on the field to deliver their full potential without redundancies.
The early warning provided by Phumus can also be extended to the population, real time maps with relevant information may be available to the public, and the emergency services will have the information to know if their current location is still within the accepted risk.

Fire fighting and mitigation is still a people’s matter, in the end the ones going in front of the blase and the ones that decide who and where to deploy still have the ultimate decision, Phumus is the enabler of the best decision.

Details

In Phumus we believe technology should have an impact in people’s life. Technology should help, be easy to comprehend and use and most of all be available when it’s most needed.

Phumus uses a combination of self built technology like it’s array of sensor devices (Patent Pending) and open source tools and frameworks.

The sensors are completely autonomous, non-maintained, non-administrated and secure using military grade encryption. Our philosophy is fire and forget devices that create a mesh of information, different kind of nodes provide information about emergency services (public of private), environment data, high value location information (ex: number of people in a nearby attraction and relate that with an event or possible event) or emergency service individuals tracking in real time.

All the sensors are made to be robust, to be deployed in the wild enduring the external conditions as temperatures bellow -30ºc and +90ºc as you have in a object exposed to the sun in an hot summer day, water and dust resistant (IP64).

The communication is handled via a custom built protocol based on radio frequency technology with a redundant link to public cloud implementations so there is not a single point  of failure. The full scope of the software and hardware is based on security, resiliency and autonomous operations. There are possibilities of integration with TETRA emergency network if desired both to feed data or to receive data and feed it to the Artificial Intelligence engine.

Apart from the patented node components Phumus also uses a multitude of Open Source components as Linux, Ambari, Graphite, MariaDB, TensorFlow, Prometheus, CockroachDB, HAproxy, nginx, and also public cloud services for computing as google compute engine and Amazon AWS.

Phumus is designed to be sole responsible for the data and information, the data is always transported in encrypted formats so no data (sensor or 3rd party) information is available outside of the scope of the Phumus platform.