Image
Intelligence artificielle

Influence croissante de l'intelligence artificielle sur l'assurance qualité

aoû. 2 2021 - 3 min

L'IA est définie comme une constellation de technologies qui permettent aux machines d'agir avec des niveaux d'intelligence plus élevés et d'imiter les capacités humaines de perception, de compréhension et d'action. À ces capacités humaines s'ajoute la capacité d'apprendre par l'expérience et de s'adapter au fil du temps. En d'autres termes, l'IA permet aux machines de percevoir leur environnement, de réfléchir et, dans certains cas, d'apprendre à agir en fonction de l'environnement et des circonstances qui le sous-tendent. Les systèmes d'IA trouvent une application de plus en plus large dans les entreprises à mesure qu'ils deviennent plus sophistiqués. 

AI also refers to the ability to deal with new situations, data, and circumstances that have not originally been anticipated. When we add experience learning, which is enhanced by the cons- tant repetition of tasks and performance measurement (machine learning), then we get an al- most autonomous system. This is something that is increasingly being deployed in robotics and in the automotive industry, in the manufacturing process itself, but also in HR departments, agriculture, pharmaceutical industry and various service sectors, with the purpose, among others, of considerable cost-savings through proper performance, targeted consumption of energy, and other resources and enhanced quality control.

GLOBAL DEMAND FOR AI 


According to the April 2021 report "Global Artificial Intelligence (AI) Industry", AI is poised to reach US$312.4 billion by the year 2027. The shifting dynamics supporting this growth makes it critical for businesses in this space to keep abreast of the changing pulse of the market.

AI is helping companies to carry out remote working, operations and management, while simultaneously improving operational efficiency. Rapid improvements in AI technology are driving increased adoption of AI-based customer service systems. AI, coupled with predictive analytics, is set to help retailers identify buying trends and quickly launch automated cam- paigns that will inspire highly targeted customers to act. Most major healthcare organisations are relying on AI-based software for their day-to-day tasks. With COVID-19 becoming a global crisis, AI became the first line of defence in using it to keep track of virus transmission. In this unprecedented situation, AI helps to reduce the work of clinicians and medical attendants. Many banks have already adopted AI-based systems to provide customer support, detect anomalies and credit card frauds. 

ARTIFICIAL INTELLIGENCE IN QUALITY CONTROL 

The use of AI has been growing across the industrial automation space for several years now, with few areas of industrial technology today remaining untouched by it. AI is changing the technologies we use to run manufacturing and processing facilities in subtle and not-so-subtle ways. 

One application with a big potential to benefit from AI is quality control. The use of smart cameras and related AI-enabled software is helping manufacturers achieve improved quality inspection at speeds, latency, and costs beyond the capabilities of human inspectors. And the timing of the arrival of these smart camera technologies is fortuitous, given the social distancing requirements of COVID-19.

While some manufacturers have been using machine vision in quality applications for many years now, the addition of deep learning-enabled quality control software represents a depar- ture from earlier machine vision technologies. Rather than having the machine vision system rely on the rules created by the expert, the AI-powered software can learn which aspects are important on its own, and create rules that determine the combinations of features that define quality products. 

IMPACT OF AI ON QUALITY ASSURANCE TESTING 


Testing services help a variety of industries meet essential regulatory and quality assurance standards, ensuring peace of mind for businesses and consumers alike. As technology advances, testing must keep pace. 

Image
Microscope MAJ

AI improves laboratory testing in two major ways. First, it accelerates the testing process, analysing thousands of data points faster than the trained human eye. Since much of the data analysis performed in laboratory testing is repetitive and time-consuming, doling out simpler tasks to AI makes analysis quicker and easier. This allows lab technicians to focus on more sensitive testing issues - such as analysing abnormal or critical samples, those of greatest concern to clients - and thereby offer more useful insights and recommendations.

Secondly, AI has the power to learn as it works, so as to recognise sample patterns more quickly and accurately over time - which is crucial for testing services. Because machine learning is not rule-based, but factors in correlations and permutations, it retains the flexibility to allow for various interpretations.

AUGMENTED LABORATORIES 


Bureau Veritas, which offers testing services at more than 400 laboratories worldwide, has been at the forefront of this AI movement with the creation and implementation of Augmented Labs. These labs use digital assistants, or Artificial Intelligence (AI), to help technicians analyse data samples. The AI program CHARLES®, designed in partnership with Microsoft and its Azure platform, serves as the forerunner for the launch of Bureau Veritas’s AI Augmented Labs.

Originally, the CHARLES® AI program—developed in Atlanta, Georgia, and named in honour of a Senior Data Analyst who devoted over forty years to working with Bureau Veritas —began monitoring oil conditions for industrial fleets. It scanned lubricant samples taken from trucks, loaders and mechanical diggers for potential risks—such as the presence of wear or conta- mination — to ensure client machinery remained in optimal working condition. With the introduction of AI, the Atlanta lab could analyse upwards of 400,000 samples each year.

Having already processed thousands of samples, CHARLES® renders reports with a high degree of confidence.  CHARLES® and similar AI programs address concerns relevant to a variety of sectors. For instance, the mining industry relies on testing services in mineral sector exploration to determine the chemical properties of their ore. “Artificial intelligence enables Bureau Veritas to predict mineralogical composition, in addition to chemical and physical properties of samples with a simple infrared spectrophotoscopy test, at just a fraction of the original cost. The process uses the same sample already being submitted to the laboratory for other routine test work. Thus AI-assisted minerals testing can save valuable time while reducing expenses.


BUREAU VERITAS ROLLING OUT AI WORLDWIDE


The success of CHARLES® is replicable, and Data Scientists are presently working to scale and deploy its AI technology, with CHARLES®  serving as the template, for the development of additional AI programs. With a strong and growing presence of Augmented Labs, Bureau Veritas will continue to provide high-quality assessment services to even more businesses, which translates to safer food, a cleaner environment and more reliably and efficiently tested consumer products.

This ambition has been promoted by the creation of the Bureau Veritas Data Lab team, whose mission is to support and develop all projects in the field that use AI. This team will benefit from the technical and commercial partnership established with Microsoft, for use of the AI resources offered by the Azure Cloud platform.

Eventually, the aim is for all tests offered by Bureau Veritas to benefit from the contribution and power of AI, in whatever business sector its clients happen to operate in, around the world.