Respond to the current lockdown challenge: Exploit your manufacturing connectivity through machine learning with ASTUTE 2020
ASTUTE 2020’s support is underpinned by recognised expertise critical to manufacturing processes. The uniqueness of ASTUTE 2020’s support is aimed at meeting the industrial need of improving the use of resources through manufacturing processes and managing the supply chain of products and services. In this article, we find out about Machine Learning and how it exploits connectivity in the manufacturing process.
What Is Machine Learning and How Does It Work?
Machine Learning (ML) is a data analytics technique that teaches computers to do what comes naturally to humans and animals: learn from experience. It is a core sub-area of Artificial Intelligence (AI) based on the idea that systems can learn to identify patterns and make decisions with minimal human intervention. For instance, AI is the ability of a digital computer or computer-controlled systems to perform tasks commonly associated with intelligent beings.
The process of learning begins with observations or data, such as examples, direct experience, or instruction, followed by identifying patterns in these to make better decisions in the future, without relying on a predetermined equation as a model. The primary aim of ML models is to allow automatic learning, improving from experience without being explicitly programmed or adjusting actions accordingly, usually with the final goal of predicting some target output or response. These algorithms are heavily based on statistics and mathematical optimisation.
When exposed to new data, these models learn, grow, change, and develop independently. In other words, with ML, computers find insightful information without being told where to look.
What Are the Opportunities and Challenges Around the Application of Machine Learning and Data Analysis?
- A new approach to Quality Control
- Predictively Identifying Issues
- Problem Source Identification Through Correlation Analysis
- Pattern Recognition for Defect Identification
- Multi-Objective Optimisation
Although ML has been immensely transformative in some fields, it can fail to deliver expected results in particular applications. Reasons for this are numerous, including but not limited to: lack of suitable data, lack of access to the data, data bias, privacy problems, badly chosen tasks and algorithms, wrong tools, lack of resources and evaluation problems.
One of the main characteristics that still differentiate human from computers is the power of Interpolation and Extrapolation, that is, respectively, the ability to extend what must have happened between observations and the ability to extend to what happens before, after, or beyond observations. In other words, the capability of reaching reasonable conclusions about observations never seen before. This is what enables human visual inspection to be more accurate than automated systems most of the time, for example.
What Are the Benefits/Impacts of Incorporating this Technology?
The use of ML and Data Analysis techniques can provide actionable insights from, usually overlooked, datasets. As the manufacturing sector advances in the era of Industry 4.0, Internet of things (IoT) and the fourth industrial revolution the value of data increases significantly. Fraud detection using pattern recognition, production costs optimisation based on evolutionary computing, premature failure detection for equipment maintenance and malfunctioning identification using classification models are some of the high-impact tools enabled by the use of ML and Data Analysis of industrial datasets.
Can ASTUTE 2020 support companies with this Technology while working from home?
It all starts with two main elements:
- A problem/challenge to be addressed.
- Data containing relevant information to the target problem.
Collecting and gathering all the necessary data is not always an easy task. If not already available, the installation of sensors may be needed to collect the required data. Depending on each specific problem, the amount of information desired for building reliable ML models can be significant.
Once the dataset is available, all the necessary analysis and modelling can be performed remotely while working from home as all the computing resources used by ASTUTE 2020 are available online to our project officers.
How Is ASTUTE 2020’s Expertise Incorporating This Technology into The Welsh Manufacturing Sector?
ASTUTE 2020 have applied AI, ML and Data Analysis successfully in collaborative projects with Welsh manufacturing industries. Projects involving the minimisation of electricity costs and feedstock through ML models and evolutionary computing are an example of this. A range of companies are currently collaborating with ASTUTE 2020 to further their knowledge and develop opportunities of incorporating ML through advanced data capture and intelligent systems into manufacturing processes, e.g. large steel producers like Celsa Steel based in Cardiff and Tata Steel based in Port Talbot, as well as smaller, innovative companies like Pro-Flow Solutions Ltd. based in Blaenau Gwent. All our industrial collaborations can be viewed here https://bit.ly/34E70cK
The uniqueness of ASTUTE 2020’s expertise across our Welsh Universities’ partnership offers manufacturers the unique opportunity to embrace future manufacturing technologies such as ML, by rethinking every aspect of their manufacturing business from a digital perspective, creating further value, improve efficiency and allow the offering of better products, processes and services to a global market.
ASTUTE 2020 can support manufacturing companies across a variety of sectors, such as aerospace, automotive, energy generation, oil and gas, medical devices, electronics, foods, etc., stimulating growth by applying advanced engineering technologies to manufacturing challenges driving cutting-edge research and innovation. ASTUTE 2020 collaborations inspire manufacturing companies to improve and streamline their manufacturing processes, manufactured products and supply chain, generating sustainable, higher-value goods and services and bringing them to a global market.
The ASTUTE 2020 operation has been part-funded by the European Regional Development Fund through the Welsh Government and the participating Higher Education Institutions.