Covering Disruptive Technology Powering Business in The Digital Age

image
Western Digital Malaysia Uses Big Data Analytics to Predict Manufacturing Defects
image
May 18, 2016 News

Challenges: As one of the largest hard disk drive suppliers in the world, Western Digital must ensure volume and efficiency in the manufacturing and distribution of its hard disk drives in order to maintain its competitive edge.  The Western Digital subsidiary that SAS partnered with produces millions of hard disk drives per year – and while the success rate is maintained at such high levels, the failure rate of even a fraction of a percentage results in the production of a million defective drives.  Therefore minimising customer losses is critical to its operations, and the company’s priority has been to minimise the distribution of such defective units.

The Hard Disk Drive (HDD) Analytics department had been using among others, JMP software for many years to perform root-cause analysis of a yield excursion. George Ng, Director of Analytics at WD wanted to move from a reactive approach to a predictive approach of doing this. In addition, it needed a solution that was able to handle the increasing amount of data its systems were producing.

WD had built a comprehensive data mart with approximately 3000 variables comprising data from various manufacturing and supply chain processes.  Due to the large amount of data, WD needed an enterprise solution and it wanted to build predictive models on that data. WD turned to SAS to provide them with a big data analytics (BDA) solution.

Big Data & Analytics Solution

SAS Asset Performance Analytics (APA) monitors equipment sensors, tags and machine-to-machine (M2M) data to identify hidden patterns that predict failures. “SAS comprehensive analytics software solution has provided us ability to do complex data analysis generating new useful analytics insights for our business. With an in-built case management system, SAS’s APA was able to give WD engineers the insights they needed to identify possible failures early in the production process and make timely decisions to avoid a yield excursion and ensure WD hard drives are of the highest quality,” says Ng.

Dr. Mark Chia, Head of Advanced Analytics Centre of Excellence at SAS Malaysia says that, “the interface of this system is designed so that engineers can perform a series of functions including the data extraction, data conversion and data analysis. In this way, all device performance indicators are monitored, so once an exception occurs with the device, the system can provide an alert to the engineer so that they can make critical decisions quickly.”

Benefits Obtained

The company’s commitment to quality improvements has ensured the highest quality product in the industry. The company’s precision in identifying a yield excursion has lowered the overall number of returned units, which in turn boosts customer loyalty and trust, and this of course has direct bearing on the company’s revenue.

Industry: Manufacturing

Client: Western Digital Malaysia

Headquartered in California, Western Digital is a pioneer in hard disk drive storage manufacturing and is one of the world’s largest hard disk drive suppliers. With SAS, WD was able to predict yield excursions and therefore reduce the losses caused by the production of defective devices.

The company’s ability succeed despite competition from new players can in part be attributed to the company’s commitment to quality.  Quality improvements are driven by complete product and component traceability on the entire life-cycle of every unit of drive manufactured, from suppliers to manufacturing, to testing, shipment and customer use.

Big Data Solutions Provider Profile:

SAS is the leader in analytics. Through innovative analytics, business intelligence and data management software and services, SAS helps customers at more than 75,000 sites make better decisions faster. Since 1976, SAS has been giving customers around the world THE POWER TO KNOW®.

This article was originally published on www.asiandatascience.com and can be viewed in full

(0)(0)

Archive