MHP unlocks big data with Microsoft Azure to effectively manage tons of grain

In the past 25 years, MHP has grown into a leading international Food and Agrotech group. Headquartered in Ukraine, its production facilities are scattered across Ukraine and the Balkans. With more than 30,000 employees, MHP supports Ukraine's food security, delivering products to communities around the globe, and even to the frontline of the war. “We are one of the companies responsible for feeding Ukraine’s population and reaching the most remote locations, especially during the war,” notes Taras Goshovskyi, Chief Information and Digital Officer at MHP.

No food to waste

But managing the movement of tons of grains and oilseeds across vast fields and storage sites demands precise orchestration. To avoid shortages or excess stock, MHP needs to strategically allocate grains across various storage locations while considering logistical constraints, market needs, processing capabilities, and future utilization within its supply chain. “We needed to plan a complex solution involving multiple interconnected parts and gather decentralized data to create models for managing grain distribution scenarios,” explains Serhii Dobrohorskyi, Deputy Chief Executive Officer, Director of Agro at MHP.

First things first. MHP needed a resilient and adaptable infrastructure. "We wanted to be prepared for different business scenarios. Secondly, we needed the flexibility to adjust our computing resources easily. For instance, to have more processing power and then scale down when not needed. That's hard to do with our on-premises hardware," recalls Taras Goshovskyi, Chief Information and Digital Officer at MHP.

So, MHP embarked on its digital transformation journey, opting for a suite of Microsoft solutions. "We've had more than 15 years of successful partnership with Microsoft, so continuing with them simply made sense,” says Taras. MHP adopted Microsoft 365 E5 and Microsoft Viva to enable remote collaboration, and migrated workloads to Microsoft Azure, transitioning to a cloud-centric approach. This shift enabled MHP to carry out complex tasks like data warehousing, big data processing, and even develop sophisticated logistical applications like object and face recognition on Azure Cognitive Services.

Making fast, precise decisions

Having set up a foundation in the cloud, MHP was ready to develop its grain balance management solution by combining complex mathematical modeling algorithms with the user-friendly interface of Microsoft Power BI. It used Azure Synapse Analytics and Azure Data Lake to collect, process, and model data for analytics.

The result is a tool that can determine the optimal storage location for 2 million tons of grains and oilseeds, considering factors like proximity to fields, storage capacities, processing capabilities, shipment logistics, and future grain utilization plans. “It’s achieved what would be impossible for a human to do—calculate the optimal placement of tons of grain for a whole year. Before, we’d just think logically and rely on our experience. Today, the grain balance management solution enables us to confirm our decisions or to see all alternative courses of action," explains Serhii.

Soon, MHP moved the solution to the cloud, bringing speed and reliability. “We can calculate more than 10 million scenario combinations for decision-making on geographical fields allocation, logistics, and storage,” elaborates Taras. “Before the solution, this took nearly two weeks and 15 people to complete. On-premises, it took us two hours. On Azure, it takes 15 minutes.”

More security, less cost

Finally, using Power Apps, MHP developed a mobile application to edit input data for modeling, allowing users to get results for different sets of circumstances within a click of a button. "Additionally, we wanted to be better guarded against more powerful cyber-attacks that were occurring lately. Using Microsoft 365 E5 and Azure, we were able to enhance our cybersecurity. For our infrastructure, we have integrated identity and access management (IAM), firewalls, and other Microsoft security tools," shares Taras Goshovskyi.

The insights MHP has gained enables it to save money on creating animal food and earn more profits from selling grains. By eliminating human error in decision-making, MHP can also plan more quickly for unexpected situations. “We have all the instruments in Azure that we need. We have powerful and insightful reports to see how the system works and can even model what will happen if we change something in the model or scenario,” highlights Taras. ”Using Azure, we’ve gained computing efficiency and save USD200,000 every year, with some scenarios saving over USD1 million annually.”

Looking ahead, MHP continues to leverage technology for more accurate and dynamic planning. "The big data scenario in agriculture is only the first step, because the biggest part of our business is producing fresh meat. We are aiming to replicate a similar approach there," Taras shares. "One day we will have a system with real-time calculations instead of forecasting the whole year ahead. This would allow us to rapidly adapt to changing circumstances, such as transport availability or operational issues, enhancing both efficiency and resilience. Many citizens rely on us for food supply, so business continuity and sustainability must come first—we must keep going.”