Skip to main content

AIRISE Use Cases - Applications in Manufacturing

stories about AI-in-manufacturing

Artificial Ingelligence (AI) already contributes to diverse challenges in manufacturing at the shop-floor level. We have collected a few use-cases which nicely show details about challenges and related solutions.

The Intelligent Planner

an application for planning production

Planning is essential for organizations and businesses: it helps establish appropriate goals, reveals weaknesses (and strengths), increases efficiency, reduces risks, and enhances decision-making. The complexity of planning and the need for re-planning have increased due to the exceptional circumstances we are experiencing, such as disruptions in the supply chain, the need to save energy consumption to minimize costs, ecological transition, and increasingly sophisticated manufacturing processes, among others. 

Key achievement of this applicaiton

  • reduction of planning effort by 20%
  • enhancement of response time by 15%

Machining in the aerospace sector

addressing issues in the production and supply chain

Planning is essential for organizations and businesses: it helps establish appropriate goals, reveals weaknesses (and strengths), increases efficiency, reduces risks, and enhances decision-making. The complexity of planning and the need for re-planning have increased due to the exceptional circumstances we are experiencing, such as disruptions in the supply chain, the need to save energy consumption to minimize costs, ecological transition, and increasingly sophisticated manufacturing processes, among others. 

The initial situation of this factory in the aerospace machining sector, is characterized by a manual process, limitations in adaptability, and the inability to simulate scenarios. Additionally, issues related to delays in parts and non-productive hours posed challenges in production management and operational efficiency.

  • Manual process: The factory relied heavily on human labour to carry out its operations. This influenced the speed, quality, and efficiency of production. 
  • Classic algorithm: The factory used traditional methods for production scheduling and task assignment, which limited its flexibility and adaptability to respond to changes in demand or operational conditions. Data obtained over two months on the 35 production machines included: 
  • 487 delayed pieces resulting from difficulties in meeting delivery deadlines or problems in production planning and execution. 
  • 2736 non-productive hours or downtime in its machines or processes caused by maintenance issues, adjustments, unplanned downtime, and other factors. 
  • Limitations in adaptability: The company had limited ability to modify its processes or equipment in an agile manner in response to changes in the environment or market needs, affecting its competitiveness. 
  • Inability to simulate scenarios hindered the factory from conducting predictive analysis and testing different production strategies before implementation, making decision-making challenging.

The Artificial Intelligence project focused on implementing a production planning tool in the factory. The Easy Planner tool is capable of dynamically generating optimal plans, adapting to each situation in the plant, and learning from previous experience. It suggests a dynamic order of jobs on machines, optimizing one or multiple Key Performance Indicators (KPIs) identified by the user, while also considering the available resources for plan execution, such as tools, raw materials, or labour.

The implementation of the AI tool resulted in significant benefits for the business. Firstly, the digitization of the process led to standardization, thus reducing the risk of operational failures

Additionally, the development of a customized metaheuristic algorithm has had a significant impact on the factory's efficiency. A 37% improvement in the number of delayed pieces has been achieved, decreasing from 487 to 305. Similarly, non-productive hours have seen an impressive 54% reduction

One of the most notable changes is the machines' ability to respond to events such as breakdowns and changes in part preferences. This has increased operational resilience and improved customer satisfaction by ensuring a more stable and predictable production flow. 

Furthermore, the planning tool has allowed for the simulation of different scenarios, leading to more efficient shift scheduling. A specific example is a machine that has experienced a 26% reduction, going from 96 to 71, demonstrating the adaptability and optimization capabilities provided by this new AI tool.

The company will roll out the solution after the validation phase.

Funding of the roll out will be realised internally.

Contact

discuss the details with the people behind this application

Luis Usatorre | tecnalia

Keen on new use-cases?

Register for our newsletter