Artificial intelligence (AI) and machine learning (ML) are being applied to all aspects of technology and maintenance is one of those. Being able to predict when a device needs service allows work to be done before a failure occurs. The latter tends to be much more expensive to fix and the consequences are more dire than the cost of servicing.
I spoke with Jake Graham, Director of Product Management for Quality and Maintenance at the Intel Saffron AI Group, about their newest product, the Saffron AI Quality and Maintenance Decision Support Suite. It brings AI and ML into the mix of industrial internet of things (IIoT) and Industry 4.0 sensors and control.
Jake Graham, Director of Product Management for Quality and Maintenance, Intel Saffron AI Group
What are some challenges that businesses face regarding maintenance and quality? How can AI solve current business needs?
When it comes to maintenance and quality concerns, business can face a major loss in profit as a result of wasted efforts and resources, customer turnover, and in some situations, safety issues. These are largely due to unplanned downtime and potential safety risks involved with system failures.
AI-based quality and maintenance solutions, a relatively new technology segment, is designed to help companies react faster when issues or defects occur in a product by providing actionable insight on how to identify and fix problems. In addition, finding trends across issues allows enterprises to form better maintenance plans and prevent problems before they cause serious damage.
When addressing quality, AI-based solutions can solve problems faster, reducing the time-to-market or time-to-resolution while increasing customer satisfaction. The solutions don’t just identify the cause of a single, isolated quality problem, but provide insights into more general issues—the trends—within design or production. As a result, businesses can build better products and, ultimately, increase customer satisfaction and revenues.
When addressing maintenance, we believe that three top drivers include the need to increase uptime, reduce risk, and cut overall maintenance.
- Increase uptime: Unplanned downtime is a significant cost driver in any industry that must maintain large inventories of capital assets. For example, an airline that experiences delayed flights due to unplanned maintenance can cost thousands of dollars each Similarly, disruptions to oil platforms and manufacturing plants can cause a major profit loss. To eliminate unplanned downtime, quality and maintenance solutions can help with planned maintenance by shortening maintenance operations windows.
- Reduce risk: While businesses take cautionary steps to comply with safety regulations and minimize the chance of accidents, the potential risk is always there. Quality and maintenance solutions can help maintenance and repair operations mitigate risks through identifying defects before they cause a greater failure to the system and put lives at risk.
- End unnecessary maintenance of assets: Due to fear of combined unplanned downtime and the risk of a catastrophe occurring, businesses tend to over-maintain most of their capital assets. By increasing uptime and mitigating risks to avoid catastrophes through seeing trends across issues, businesses can reduce wasted efforts in over-maintaining assets.
How is Intel addressing these challenges?
Intel recently announced the Intel Saffron™ AI Quality and Maintenance Decision Support Suite, a suite of AI-powered software applications using associative memory learning and reasoning solutions, to facilitate faster issue resolution in manufacturing, aerospace, and software.
Can you tell me more about the new Intel Saffron AI product and how it works?
The Intel Saffron AI Quality and Maintenance Decision Support Suite is comprised of two software applications:
- Similarity Advisor finds the closest match to the issue under review, across both resolved and open cases, identifying paths to resolution from previous cases and surfacing duplicates to reduce backlogs.
- Classification Advisor automatically classifies work issues into pre-set categories, regulator mandated or self-defined, speeding up and increasing reporting accuracy while improving operations planning.
It works through Intel Saffron AI’s capability to simulate a human’s natural ability to learn, remember and reason. It uses associative memory learning and reasoning to analyze structured and unstructured text data sets to find hidden patterns, trends, and similarities. It surfaces tribal knowledge—insights about previous issues, how they were resolved, who resolved them, and what information was needed—in a transparent way through an easy-to-use interface, so that human operators can make better-informed decisions.
What types of industries does it help, and what are some benefits that each industry can experience after using it?
The Saffron AI product helps facilitate faster issue resolution for complex manufacturing, aerospace, and software development. These industries can benefit from:
- Decrease issue backlogs
- Improve issue resolution efficiency
- Boost product quality
- Optimize supply chain
- Eliminate duplicate work
- Increase aircraft uptime
- Faster issue resolution time
- Optimize supply chain and inventory
- Help meet key FAA regulatory requirements
- Lower risk and liability
- Reduce cost by eliminating duplicated bugs
- Propagate tribal knowledge across teams and regions
- Improve product quality
- Faster time to market
Can you share an example of a customer that’s using the new Intel Saffron AI product? How has it helped impact their business?
Accenture, a global professional services company, was facing an increasing complexity of software on one hand and an exponential growth of connected products and devices on the other hand. In the meantime, test engineers require access to domain, analytics, and data-management tools that are more sophisticated than traditional testing platforms. By using the Intel Saffron AI solution that identified and flagged similar bugs, Accenture is able to consolidate defects and reduce 30% to 50% duplicated engineering effort in software testing.
How does the new Intel Saffron AI compare to other offerings?
The Intel Saffron AI Quality and Maintenance Decision Support Suite differentiates from other market offerings through its ability to take any type of data, sparse or incomplete, as well as structured and unstructured text data from various sources, to provide explainable recommendations and results. It also doesn’t require traditional training models and continuously learns from user feedback and new data inputs.