In the mid of last decade, the terms Artificial Intelligence (AI) and Machine learning (ML) gained popularity. Initially there was a lot of insecurity about the concepts. But the technology prevailed, and Artificial intelligence evolved. Today, every ‘Tech’ literate person is using Artificial intelligence in some way or other. AI has entered all realms of businesses and the advantages were immediately known – which made adaption easier and faster.
So, what is the difference between AI and ML?
AI is a broader concept, with the objective of creating intelligent machines with the capability to emulate human thinking. Whereas ML is a subset of AI, which would allow to build ‘Learning machines’ i.e. machines which learn, understand and deduce based on data available and thereby evolve.
A further subset of ML is Adaptive learning. Adaptive learning merges the previous rule-based, simple machine learning, and deep learning approaches to machine intelligence. With Adaptive Learning, machines can understand new set of data and process the same for improved functioning. Adaptive learning requires the least amount of human intervention.
The advantages of adaptive learning are multi fold:
- Easily identifies new trends and patterns by reviewing large amounts of data.
- There is no need to babysit a project, with ADL capabilities self-learning becomes the norm.
- Continuous improvement is possible.
- Processing multi-dimensional and variety of data is possible within fraction of seconds.
What is Product Grading and What Adaptive Learning has to do with it?
Grading is the process of separating products based on certain pre-defined conditions, which might be related to size, shape, colour, type, weight, etc. The most significant advantage of efficient grading is the pricing of a product – higher the grade better the prices and more the profits. It also helps in optimizing production processes to achieve a specific grade. Traditionally grading is done manually, which is prone to subjectivity and overlooks- at times, which can be expensive or back-breaking.
General grading could be as follows:
|Grade A||Grade B||Grade C||Grade D||Salvage|
|These mobile phones are just like new, with minimal signs of usage and fully functional||There is a possibility of the phone having numerous minor scratches on the screen.||This includes mobile phones with signs of heavy use, lost parts, dents||Functions may show intermittent problems but still allow for essential phone communication.||Phones require a lot more work to make them operational. They do not power up, have dents, or are broken|
Let’s take an example
In many developed countries, buying Carrier Phones is a norm; carrier phones are those phones which are sold over the counter by telecom operators. There are two processes that would be followed-forward logistics and reverse logistics. Forward logistics is the process when a customer buys a phone and it is delivered to him. Reverse logistics is when a customer, after buying a phone, returns it back to the seller. Further, reverse logistics involves remanufacturing, refurbishing and reuse of products and materials.
A question arises, what happens to these returned phones?
A simple answer could be – they get discarded. A thought-through solution is- they are recycled and put back into the market. Phones returned to the carriers and sent to the OEM’s which then recycle them . But recycling a phone is not as simple as it seems. – Once sold, Phones return with minor to major damages. Damages can range from invisible scratches to cracks, digs or dents or even discoloration etc. Indeed, the OEM has to refurbish the phone then and send it back to the market as a new product. But for the returned phones to get repaired, the OEM’s need to know the volume of damage, or – the grade of damage inflicted
And hence, enter AI and Adaptive learning…
AI’s adaptive learning element, helps OEMs to grade and determine the condition of damage. This way, an OEM can assign the needed resources for refurbishment. For a precise ROI, it’s important to know what expenses are required and the revenue impact. With AI-based machines, a report can be generated on the extent of scratches and variations in damage. With Adaptive learning, the grading machine can further enhance its learning and – generate more granular reports for further processing by the OEM.
We at Griffyn Robotech , have recognized the importance of AI and Adaptive learning. We are one of the leaders in developing AI-based cosmetic grading machines for mobile phones and tablets. Our product DEEPSIGHT®, eliminates the need for manual grading, which in return eliminates subjectivity and costly overlooks. It helps decision making faster and effective. Installing DEEPSIGHT® for your reverse logistics support will help OEMs plan and manage resources for refurbishments effectively and help improve bottom lines too.