Embedding KNIME in a Manufacturing Environment

Author:Brendan Doherty, Seagate Technlogy

In this blog post I will discuss some of the processes and steps that were taken on the journey to embed KNIME in a High Volume Manufacturing Environment within Seagate Technology.

Seagate Technology are one of the world's largest manufacturers of electronic data storage technologies and solutions. Seagate Technology creates products and services that include network attached storage, high performance computing, data protection appliances, internal hard drives, backup and recovery services, flash storage, and related solutions. They are a vertically integrated company and have manufacturing plants based in many locations worldwide. The read/write heads for the HDDs are manufactured in two locations, one of which is in Derry City, Northern Ireland. These devices are highly complex, have a long manufacturing cycle time, and generate a lot of data during their fabrication. The plant has many different groups located at the site all of which use data from a wide variety of sources on a daily basis.

The Operations Technology Advanced Analytics Group (OTAAG) within Seagate Technology recognized the potential for the application of KNIME Analytics Platform to the data centric world in which the employees of Seagate work, in order to help to make better data driven decisions and also automate many repetitive manual data tasks at a variety of Seagate sites around the world (US, Asia, Europe). After having used KNIME for a range of projects for around about a year I attended the KNIME Spring Summit in Berlin. I was enthused from this experience and from seeing how other companies were using KNIME, so I set off to develop a pathway to advocate the use of KNIME and embed it within the plant and other Seagate sites that use KNIME. Initially there were only a few users all based within the OTAAG group, but now after a lot of hard work and persistence all of the major and many of the smaller groups in the factory are showing varying levels of usage of KNIME for a wide variety of tasks. My aim now is to encourage the cross pollination and synergy between groups, where KNIME users exchange concepts and knowledge, which is all to the benefit of them and the company (see Fig. 1).

Requirement: a tool for all levels of users to achieve data engineering needs

The diagram in Figure 2, below, highlights some of the steps that I implemented to enable the uptake of KNIME in the plant. Prior to this point there was a definite interest and appetite to learn more about Data Engineering and Data Science across a range of groups, but due to varying levels of user experience and ability this ambition had not been achieved.

Having programmed in many languages in the past before using KNIME, I immediately saw the opportunity to advocate the use of KNIME as a tool for all levels of users to engage with and enable them to achieve their data engineering needs.

The first part of the pathway was to deliver hands on training using KNIME to many waves of users across a range of groups. By using specific factory orientated examples, which catered to all levels of users, people could immediately see the benefits and opportunities of using KNIME Analytics Platform. The training was pitched at a pace and level so as non-native programmers could easily follow and understand the examples and not feel overwhelmed by the experience.

Once the training was complete I encouraged and helped people to then complete a use case based on their own job function in order to cement learning. An enthusiastic approach is critical here and I worked with many users on a 1:1 basis in order to get projects kicked off and driven to completion. In the long run all this work benefited both the users and the company.

Newly learned skill set quickly employed

I found that most users are able to get up and running quickly after the hands on training, in many cases within a week or two. One interesting example of this was a project from an intern who had limited programming skills when they started to work in the company. The user was able to get up to speed in the use of KNIME very quickly and then saw an opportunity when they could employ their newly learned skill set. The intern created a solution, which highlighted tools that were not following complex dispatch system rules, thereby having some knock on effects in the manufacturing line. By highlighting affected tools and taking corrective action the speed at which material was moved improved and therefore the manufacturing cycle time was decreased. In a short space of time the company was able to get a return on investment from the intern, and the intern also achieved a return on investment by developing a new skill set and creating a real world solution along with gaining invaluable experience in the process.

Tool deployed to automate time-intensive task

Another good example of how an employee was able to quickly deploy KNIME to create a business solution is from the Engineering department. Once the engineer had completed some hands on KNIME training they were able to see the potential to automate a very time intensive task for their group. This task involved a rotating list of engineers having to trawl through a variety of Google documents and databases on a daily basis in order to put together a document that identified which products were on hold in the manufacturing line. Our factory is highly integrated with Google tools, and so the Google API integration with KNIME¹ has been of real benefit in many cases. This automated solution now saves around 90 hours of engineering time per month. It also reduces cycle time by letting staff in this department tackle products in a hold state as soon as they come into the office each day, instead of having to wait on a report to be compiled and sent out.

Early adopter use case sessions & Citizen Data Science program improve engagement

A shared area and Google+ site where KNIME users could communicate and share useful documents and suggestions were setup, all in a bid to improve engagement and communication with the ever growing KNIME community in the factory.

Once I felt we had enough critical mass of projects and users I then facilitated early adopter use case sessions whereby users who have benefited from using KNIME presented to their colleagues on the solutions they developed. This word of mouth advertising helped gather traction from people who had previously not engaged in using KNIME.

At KNIME Spring Summit 2018 some of my colleagues (Allan Luk and Eric Lin) presented on the Citizen Data Science program, which they were rolling out in Seagate Technology, an important element of this is using KNIME for Guided Analytics. This initiative really dovetailed well with embedding KNIME in the factory.

Empowering people to discover and apply Machine Learning

Now that there are many people onsite using KNIME, this has also worked well with the Citizen Data Science program. It has empowered people who previously may have not considered the discovery and application of machine learning techniques to dip their toes into the world of Data Science!

About the Author:

Brendan Doherty is a Data Scientist within the Operations Technology Advanced Analytics Group (OTAAG) in Seagate Technology and is based in their site in Derry, Ireland. With over 12 years of experience in Data Analytics, Business Intelligence and Automation, he works to deliver innovative solutions, along with training and evangelizing KNIME within Seagate Technology.

Seagate is the global leader in data storage solutions, developing amazing products that enable people and businesses around the world to create, share and preserve their most critical memories and business data.