Scientific
ACT Group Inc. provides software development and algorithms implementation consulting services to companies and organizations where extensive scientific mathematics knowledge is required.
Forests in Canada cover about 49 percent of the country's land area.
In Ontario, various forest types cover about 85 percent of the province. It is one the most valuable natural resource in Canada and an important role in the economy and natural ecosystem.
A wildfire, also known as Forest Fire is an uncontrolled fire often occurring in wildland areas. Fires cause millions of dollars' worth of damage, destroy houses and agricultural resources sometimes forcing the evacuation of entire regions or cities.
There are many causes to forest fires and predicting the fire occurrence and spread has always been a major concern.
Our company was involved in development of the stochastic disturbance simulation and real-time visualization model of the fire spread natural phenomena.
The application examines a variety of probability conditions and attempts to predict the fire spread mechanism over time.
The stochastic forest fire spread model operates on a geographical terrain region presented in a Terrain Analysis System(TAS) format.
TAS input file carries the vegetation and hydrography information about the study region. The model uses the input data to create stochastic disturbance patterns on the landscape.
The forest fire spread model is a space-time Markov process which generates a stochastic simulation of growth patterns by utilizing specific parameters of the studied region.
This unique approach differs from other pattern generation tools that work on a deterministic basis.
Implementation also includes a mechanism for fire spotting, in which burning firebrands are carried aloft ahead of fire line, igniting new fires.
For more information please refer to the Forests, Fires and Stochastic Modelling page.
The system is designed using a client-server architecture. The server component is responsible for the computational engine and algorithm implementation. It's written in C for better performance and easy portability. The client shell is currently available for the Microsoft Windows platform only. It provides Graphical User Interface for data input, output, visualization and reporting.
The software package is capable of examining a variety of probabilistic conditions for the fire spread mechanism. It has over a dozen parameters that can affect the fire spread. The implementation allows real-time visualization of the fire spread both on a terrain region and contour chart. It illustrates the fire spread over time with and without spotting. It has a number of working scenarios that are randomly generated as well as working templates with real terrain and historical data.
The Natural Flow management system is a Multi-Phase Project implemented by the ACT Group Inc. The purpose of the project was to convert FORTRAN Programs used for determining natural flow for apportionment purposes on water systems into a modern, robust, and high-performance application.
The system includes mathematical calculations, database records maintenance and a user-friendly interface giving the opportunity to easily choose and navigate through various modules. After reviewing these project requirements a modular approach using the Model –View-Controller (MVC) application architecture was chosen for this task. MVC is a widely used architectural pattern that offers a clear separation of the data access and business logic from the data presentation and user interface. Our legacy specialists handled the migration of data, the mathematical experts developed a model and the user experience team designed a modern, intuitive interface.