Soil analysis is a necessary tool in crop development. In order to make optimal, fully informed decisions on crop, soil and nutrient management, farmers must have a complete picture of the chemical, physical and biological status of their soil.
However, conventional methods of data management for soil analysis are challenging due to their largely manual and disparate nature: data is not collected or organized in an automated way, requiring significant time and resource investment by laboratory personnel. In response to these challenges, soil analysis – and, in fact, the agricultural industry as a whole – is experiencing a push towards digitalization. Digitalization offers new ways to integrate advanced technologies and data strategies into operations, and create an environment in which information is used to its full potential to optimize business processes, drive revenue, and create new opportunities for innovation.
Digitalization is an especially important consideration for the agricultural industry given the ever-increasing nutritional needs of a growing global population. To overcome production difficulties and help meet these needs, laboratories can utilize advanced laboratory information management systems (LIMS) and artificial intelligence (AI) technologies to centralize their data and optimize crop output.
The Challenges of Conventional Data Management
Traditionally, workflows for managing soil analysis data rely heavily upon manual processes – and encounter a number of challenges associated with this lack of automation. These methods require extensive and time-consuming use of software such as Excel, and data must be captured, recorded, reviewed, analyzed and organized manually. Such processes are highly time- and resource-intensive, and tie up laboratory expertise that could be used more productively elsewhere.
Manual workflows are especially complex for soil analysis laboratories receiving and processing a large number of samples, as delays and bottlenecks can significantly limit throughput and performance. Distributed networks also experience additional hurdles; for laboratory setups that operate across numerous different regions and countries, manual and disparate data management processes stymie collaboration. In agriculture, being able to share soil analysis information quickly and reliably between regions is essential to drive best practice, advance research, and enable efficient, informed decision-making.
Modern LIMS and other enterprise systems, such as platforms built on AI and business intelligence (BI), enable new levels of connectivity that benefit every aspect of an organization’s data management and analysis. Using advanced LIMS, sample and process data can be collated to create a network of collective agricultural information that is easily accessible and traceable. This traceability is key for applications like soil analysis; it enables laboratories to identify issues in crop growth, and determine the root cause so farmers can efficiently and effectively address the problem at their location.
Advances in Integrated, intelligent LIMS
Innovative LIMS and AI technologies hold high value in the agricultural industry. Intelligent technologies, such as these, will play a key role in enabling agriculture to provide the output needed by the global population, which is estimated to reach 9 billion by 2050. The skyrocketing population has driven significant growth for AI in the agriculture market, which is forecasted to reach $2.6 billion by 2025 [1]. To meet demand, the agricultural sector must drastically increase production – and innovative technologies are anticipated to underpin and facilitate this change.
While data and information management systems must be robust and reliable in order to keep data secure, they must also be flexible enough to adapt and comply to the evolving workflows and regulatory guidelines of the agricultural sector, and remain open to opportunities for optimization and connectivity.
Integrated systems, such as Thermo Scientific SampleManager LIMS software, enable robust-yet-versatile working, and offer numerous benefits over traditional, and largely manual, methods of data management. Integrated LIMS hold data in a centralized, secure, traceable database. Processes – including data transfer – are automated, eliminating time delays and removing the need for error-prone paper documentation or manual intervention. Modern LIMS are flexible and easily integrated with existing laboratory instruments, offering valuable interoperability with AI and BI technologies and applications.
Applying LIMS to Colombian Soil Analysis with Agrosavia
Achieving a connected, centralized way of working is essential for distributed enterprises. The value of intelligent, integrated data management is well demonstrated by Agrosavia (Colombian Corporation for Agricultural Research). Agrosavia, a scientific and technical public entity of mixed non-profit participation, operates across various locations throughout Colombia to engage and support the development of the country's agricultural sector.
In 2018, Agrosavia received funding from the Colombian government through the Ministry of Information Technologies and Communications (Ministerio de Tecnologías de la Información y Comunicaciones) to optimize the soil analysis process. The initiative aimed to give farmers access to site-specific recommendations on crop fertilization and management prior to planting season, to enable early and informed decision-making on crop management. To achieve this objective, Agrosavia sought a single solution to connect not only the instruments in its laboratories, but also the various locations across its network, to create a collaborative working relationship with local farmers.
Agrosavia implemented SampleManager LIMS software and an AI-enabled user portal. The LIMS allows the organization to efficiently acquire, store and manage information across its laboratory network, and centralizes their databases to be more easily accessible to the farming community. Data is available via a portal, allowing farmers to register and track the progress of their samples, and view results when notified of their completion by email or text. The portal is equipped with an AI-based predictive information management system, which is able to develop fertilization plans and analyze large amounts of data to generate recommendations. The AI system was trained with approximately 10,000 fertilization recommendations from Agrosavia agronomists and, as a result, is capable of making predictions about 200 different types of crop. Through ongoing human validation, the AI system learns to continually improve its predictions, resulting in increased accuracy over time.
Agrosavia’s LIMS-enabled workflow neatly integrates different laboratories working across various scientific specialties and locations. This connectivity improves the management and traceability of both analytical information and the samples themselves, reducing processing time and increasing overall workflow efficiency in the laboratory. Together, the LIMS software and AI system have doubled the number of recommendations Agrosavia agronomists are able to complete in a day, vastly improving the organization's efficiency and productivity. An important element of the initiative is workflow automation; by using the LIMS, Agrosavia is able to move away from manual, Excel-based processes that required lengthy cooperation across numerous departments to generate a single report. Now, a single person can log on to the system, download a specific project or batch, run the analysis, and have their results automatically captured by the instruments and LIMS and logged on the centralized, accessible database.
Bringing Accuracy, Efficiency and Accessibility to Soil Analysis
LIMS and AI technologies promise to play a key role in enabling soil analysis laboratories to improve crop output. When used in conjunction with AI technology solutions to record, monitor, analyze and share data, LIMS can provide data-driven recommendations to improve fertilization plans, and bring increased production at a lower cost with optimal fertilizer use. These benefits are possible even across a complex, nationwide laboratory network that collaborates directly with numerous local farmers and partnering courier services for sample delivery and transport, as demonstrated in Colombia by Agrosavia. A comprehensive, forward-thinking data and laboratory information management strategy enables soil analysis laboratories to work at their full potential, and to open up new opportunities for future innovation. Such strategies offer an integrated, automated, forward-facing solution for the agricultural sector, and streamline laboratory workflows to bring the improved efficiency and productivity necessary to optimize crop output and enhance competitiveness.
Author Notes: Ajay Shrestha is a manager for Thermo Fisher Scientific's technical & development operations division.
References
[1] MarketsandMarkets (2020) Artificial Intelligence in Agriculture Market by Technology (Machine Learning, Computer Vision, and Predictive Analytics), Offering (Software, Hardware, AI-as-a-Service, and Services), Application, and Geography - Global Forecast to 2026. https://www.marketsandmarkets.com/Market-Reports/ai-in-agriculture-market-159957009.html