WHITELOGO

Blog

Focus on

DIAMOND TECHNOLOGY

Artificial intelligence in diamond technology

ARTIFICIAL INTELLIGENCE LIKELY TO GENERATE
NEXT GREAT WAVE IN DIAMOND TECHNOLOGY

Artificial Intelligence (AI) is likely to spur the next great technological wave in the diamond industry, quite possibly raising the capability of automated systems to a level that once would have been unimaginable.

Indeed, one of the reasons that automation came relatively late to the diamond business, and to some degree still remains relatively limited in its application, is because diamond processing cannot be approached in a standardized manner. The automobile industry, for example, has developed incredibly sophisticated robotic systems, which nonetheless turn out car after car, each one practically identical to the one that preceded it. But every rough diamond has individual characteristics, which require that it be examined, analyzed and processed individually. This invariably has demanded a measure of human intervention.

One of the breakthroughs in DIAMOND MANUFACTURING occurred in 1990s and early 2000s, when a group of mainly Israeli developers developed a model that revolutionized the manufacturing process. 

Until then, the belief was that automated systems which mimic the manual process of cutting and polishing diamonds was the way to go, reducing the cost of labor by having machines work as a pace and a degree of exactness far in excess of what human workers were capable of. What the Israelis realized was that the most critical part of the process was the planning and decision-making stage, during which a plan of action for each individual stone would be formulated. Once that was programmed in, automation could do its job.

By splitting the planning and processing stages, technology took a major step forward. Not only did it benefit from the speed and precision of automated mechanics, but carefully thought-out decisions during the planning stage saw higher yields from the rough stones, and polished goods that were better suited to what the market demanded.

MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE

But while the most sophisticated planning systems, some of which are able to delve deep into the stone, providing the diamond manufacturers with a range of manufacturing options, it is up to the human operator to decide what path should be selected. There may literally be hundreds of considerations, ranging from the shapes or sizes of stones that could be extracted from a single piece of rough, to the quality of their color and clarity, and the likelihood that the combination selected would provide optimal economic benefit at that specific point in time.  

It is here where AI can make a critical difference. Machine learning systems involve algorithms that enable the computer to identify patterns from large amounts of data fed into it, and then through a series of trials and errors learning to make predictions. It is not greatly different to the way in which humans learn, but a computer can do it for hours on end, without taking a break or getting distracted.

Artificial Intelligence actually goes beyond basic machine learning, because it involves reflection and the optimization of decision-making. Like humans, it adapts what it knows intuitively by observing the surrounding environment, and then making an intelligent decision.

MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE

A screen shot from Sarine’s Advisor system, which today the world’s most widely used rough planning software. Integrates internal inclusion scanning information and geometrical 3D analyses, and provides multiple planning solutions, the addition of artificial intelligence would enable commercial decisions to be made with a minimum of human intervention.

How could that work in the diamond business? Using existing computer-aided design (CAD) systems, most of which involve sophisticated optics and imaging technology, the computer will be able to suggest a range of manufacturing options that are to likely provide optimal benefit. But it will then refine the process by a considering up to date market data, automatically reaching a decision based on more information than any human reasonably could absorb. And that decision could change from day to day, depending on what is happening in the marketplace.

RUDIMENTARY AI-BASED SYSTEMS ALREADY AVAILABLE

As a relatively small industry, the diamond sector is unlikely to attract the types of investment that would place it at the cutting edge of AI. But as has been the case of with other technological advancements in the sector, it is benefitting from research being done in other industries. AI is likely to transform entire disciplines, including medicine where it is being adapted to make diagnostic decisions that one could only be made by human doctors, engineering, accountancy, finance and investment, insurance and more.  

And initial work is already being done. Sarine Technologies, the Israeli company that led the field in the development of independent decision-making systems in the early 1990s, has already created a fully automated diamond grading lab, using devices that operating using AI.

More recently, another Israeli venture, DiaCam360, launched a project that will lead to the automated identification of a diamond’s color and clarity, based on a proprietary database of hundreds of thousands of images of diamonds. All these images were analyzed by a deep-learning and AI platform developed by Matrix, an Israel-based software developer. The diamonds in the database were graded by GIA and have GIA diamond grading reports.

 

\

SEARCH DIAMONDS
Search