Monitor and improve the efficiency of production processes (challenge retrieved from the literature).
Challenge:
Energy prices are continuing to rise higher and higher. It is at the point where energy costs are comparable to the cost of raw materials. The company is a medium-sized enterprise consisting of several small food product manufacturers. They are Farm management is a challenging task from multiple perspectives. Agricultural processes are far from being optimized and they require manual procedures to be scheduled and to obtain useful data. Also, the need for improving sustainability and clever waste and energy management is becoming more and more urgent.faced with the challenge of needing to reduce their energy costs while maintaining production results.
Main Requirements:
- Data acquisition from all stages of cultivation, sowing and harvesting in order to support the decision making process;
- Avoid unnecessary waste by optimizing the amount of energy and water needed;
- Support the adoption of novel technologies and agricultural techniques that require more precise interventions;
- Improve the traceability of the supply chain.
Industry Sector:
Agriculture
Challenge classification:
Limiting manual checking and data acquisition to minimize human errors and have real-time process monitoring and optimization. Waste management based on real-time tracking and demand forecasting. Supply chain transparency and reliability improvement.
Time for Project Completion:
36 months.
____
Preliminary Analysis:
The taxonomy allowed us to identify the technological requirements needed for automatically acquiring knowledge and data. The company requires the deployment of IoT devices and UAVs to collect different kinds of data. The differences between these instruments relies on the level of expertise required and the type of data they stream. This data needs to be streamed to the Cloud via appropriate communication protocols and properly stored (by using standard relational databases or specialized databases for time series), visualized with business intelligence and data visualization methods, and used for extracting relevant knowledge thanks to Data Analytics Platforms and Artificial Intelligence algorithms. These two methods differ for the complexity of the predictions and the level of knowledge needed to adapt them to particular scenarios.
A possible solution is to leverage the latest evolution of IoT devices, data analysis and visualization tools and artificial intelligence and machine learning algorithms to have a clearer view of the whole farming process. By using and visualizing the data we can optimize the various phases both from a cost and an environmental point of view. To add more features to the solution, we can use the digital twins paradigm to monitor all the steps of the supply chain, maybe combining it with the Blockchain technology.
____
Solution Summary:
The schema of a possible solution is represented in the figure above. As shown, the data collection phase is fundamental for all the following phases and possible future developments. This preliminary stage would require the adoption of various types of devices sending data to the Cloud. Then, the data can be analyzed and by adopting the latest AI and Machine Learning models to address predictive tasks as well as other optimization algorithms to optimize the cultivation phases, and visualized on clear and user-friendly interfaces.
Additional Benefits:
- Enhanced Data-Driven Decision Making: The solution enables data-driven decision making, leading to improved efficiency and optimized agricultural practices.
- Improved Resource Allocation: Accurate forecasting and optimization capabilities allow for effective resource allocation, reducing costs and environmental impact.
- Potential for Commercialization and Investor Attraction: The solutions’ innovative nature and transparency can attract investors and enhance the company’s market position.
Possible Difficulties:
- Technical Complexity: Implementing a comprehensive solution involving various technologies and components can pose technical challenges.
- Workforce Skill Gap: Finding and hiring highly specialized professionals with the required expertise may be difficult due to skill shortages.
- Scalability and Compatibility: Ensuring the solution is scalable and com-patible with future advancements can be challenging.
Sources:
[1] Senthil Kumar Swami Durai, Mary Divya Shamili, “Smart farming us-ing Machine Learning and Deep Learning techniques” (2022), Decision Analytics Journal, Volume 3, 100041,ISSN 2772-6622.
[2] Digiteum, «Impact of Internet of Things (IoT) on Supply Chain Management», Accessed 24 August 2023. https://www.digiteum.com/iot-supply-chain/.
[3] Sharmilah T., Yahya M., Fuzi K.M., Pauzi M., Shirlynda Z., Faris N.A., “Development of temperature monitoring towards Industry 4.0” (2019),AIP Conference Proceedings, 2129, art. no. 020034.
[4] Bazaz S.M., Lohtander M., Varis J., “5-dimensional definition for a manufacturing digital twin” (2019), Procedia Manufacturing, 38, pp. 1705 -1712.