Smart Decisions for Future Cities.

The platform that transforms complex data into effective public policies and sustainable economic growth.

About Growa

We're Growa, a group of Economics students from the Autonomous University of Sinaloa in Mazatlán, Mexico. We tackle global challenges with solutions based on science and technology. Through innovative and collaborative initiatives, we drive real and mindful progress for our community.

The Problem: Governing in a Volatile World

Today's cities and governments face unprecedented challenges. Climate volatility threatens food security, rapid urbanization strains healthcare systems, and economic uncertainty demands more efficient resource management than ever before.

Making long-term strategic decisions based on intuition alone is no longer enough. A new generation of tools is needed to interpret the complexity of the present and model a more prosperous and secure future.

Our Solution

Introducing Growa-gov: Data Transformed into Decisions

Growa-gov is an advanced analytics platform designed for the public sector. Our mission is to empower planners, government departments, and businesses to make proactive, evidence-based decisions.

We translate geostatistical, economic, and climate data into clear answers to your most critical questions. We move beyond static reports to offer you a dynamic intelligence ecosystem that anticipates risks, identifies opportunities, and quantifies the impact of your policies.

We have focused our platform on the modules of Food Security, Public Health, and Urban Development because they are not isolated areas; they are the interconnected pillars of sustainable economic development. A predictable and resilient agricultural sector stabilizes prices and anchors the local economy. A proactive public health system reduces government costs and maintains a productive workforce. Smart urban planning optimizes infrastructure, attracts investment, and raises the quality of life. By strengthening these three pillars, we lay the groundwork for robust and lasting economic growth.

Our process transforms the noise of data into a clear signal for action. First, we create and consolidate robust databases, integrating diverse information sources. Next, we apply advanced analytics with machine learning and econometric models to geoscientific data to uncover patterns and predict trends. With these findings, we prepare the information in an accessible format: intuitive dashboards, risk maps, and executive reports. Finally, this entire process culminates in action: we equip leaders with the evidence needed to execute public policies with confidence and precision.

Strategic Modules

Our platform tackles the complex reality of cities through three strategic pillars: Food Security, Public Health, and Urban Development. We selected these areas because a city's resilience isn't built in silos; the same climate volatility that threatens food security is what, combined with urban expansion, creates the conditions for public health crises. This integrated approach is precisely what inspired the NASA challenge "Data Pathways to Healthy Cities and Human Settlements," which seeks to align the well-being of people and the environment through data. This synergy allows us to transform complex data into high-impact public policies, generating a cycle of robust economic growth that strengthens cities in a volatile world.

Food Security and Economic Analysis

Anticipate inflationary pressures and stabilize markets by transforming climate uncertainty into an economic advantage. Our unique approach not only predicts agricultural yields but also quantifies the exact cost of climate events. With this intelligence, you can analyze how supply shocks will affect prices and your region's comparative advantage, enabling the creation of proactive and effective policies.

To achieve this, we have developed an advanced analytics platform that projects agricultural production capacity. The core of our technology lies in the synthetic indices we have created, such as the "Prime" Condition Index (PCI) and the Climate Stress Synergy Index (CSSI), designed to evaluate climate quality and risk in productive terms. These metrics allow us to go beyond simple prediction to quantify the precise economic cost of climate events and analyze how supply impacts will affect price formation and a region's comparative advantage.

Farmer uses ai technology to monitor crops.

Predictive Public Health

We anticipate viral disease outbreaks, starting with Dengue, to optimize the management of health resources and reduce hospital saturation, as well as to promote more comprehensive healthcare.

Epidemiological Models: We integrate variables from urbanization, climate change, and demographic factors to identify risk "hotspots" before they turn into crises.
Resource Planning: Our platform helps direct prevention campaigns, fumigation efforts, and medical resources to the areas that will need them most

We offer the tools to intelligently plan city growth, ensuring quality of life and long-term sustainability.

Expansion Analysis: We model urban growth to optimize infrastructure for transportation, housing, and public services.
Climate Resilience: We identify areas vulnerable to climate risks like heatwaves or floods, allowing for the design of more resilient infrastructure and the protection of communities.

Sustainable Urban Development

man in white jacket wearing blue goggles

timelapse photography of vehicles and buildings

Our Technology

For predictive analysis, our platform is built on a hybrid deep learning architecture designed specifically for multivariate time series. We use a 1D Convolutional Neural Network (Conv1D) that acts as a feature extraction system to identify recurring "motifs" in climate data. Subsequently, the abstract feature sequence generated is processed by a Long Short-Term Memory (LSTM) network, whose function is to model the long-term causal dependencies and context between these events. This hierarchical synergy allows us to interpret agro-climatic, public health, and urbanization dynamics with superior depth and precision.

It allows you to:

Detect local and recurrent climate patterns, or "motifs," in short time windows, such as the signature of a drought or a cold front.
Transform raw climate data into a high-level feature map, turning complexity into meaningful "semantic events."
Recognize these critical events with the same effectiveness regardless of the time of year they occur, thanks to its translation invariance property.
Increase computational efficiency and drastically reduce the risk of model overfitting through parameter sharing.

Detecting Climate "Tipping Points"

The fundamental importance of our first technological layer, a 1D Convolutional Neural Network (Conv1D), lies in its ability to act as a perception system that abstracts the complexity of raw climate data into actionable insights. Operating as a filter bank that slides across the time series, this layer simultaneously analyzes all climate variables in short windows to detect recurring local patterns or "motifs," such as the signature of a developing drought. Its function transcends simple identification; it transforms the time series into a new, high-level feature map, moving from raw data to semantic events.

It allows you to:

Divide the analysis into two specialized stages: perception (CNN) and reasoning (LSTM), mimicking the way a human expert processes information.
Selectively remember critical information from early stages (e.g., planting) and connect it with results in later stages (e.g., harvest), overcoming the limitations of short-term memory.
Understand the "narrative" of the season, learning how the sequence and order of climate events (the "grammar of the climate") affect the final outcome.

Hybrid Architecture of Convolutional and Recurrent Neural Networks (CNN-LSTM)

The second technology is the hybrid CNN-LSTM architecture, a complete system that integrates the first. Its innovation lies in its two-stage structure: first, the CNN acts as a perception system that abstracts raw data into meaningful "motifs." Subsequently, the LSTM layer—the core of this second technology—uses its long-term memory and gates to interpret the sequence of these motifs. Its function is to model the causal narrative of the season, allowing for robust predictions that consider the complete temporal context.

Our Foundations

Our team and our vision would not be what they are today without the guidance of our mentors. We would like to introduce you to the people who inspired us:

Lic. Miriam Nayelly Munguía Lizárraga

She provides the essential bridge between theoretical concepts and community action. Her hands-on guidance in social projects and workshops ensures our work is deeply rooted in real-world educational and social needs.

Dr. Ulises Suárez Estavillo

Dr. Suarez is a leading voice in Social Sciences, Economics, and Politics who is dedicated to unlocking his students' potential, mentoring them to become creative and impactful agents of change.

Innovation with Impact

We believe that knowledge is the engine of change. We bring science and innovation workshops to communities to equip them with the necessary tools to build and lead their own environmental solutions.