Do You Need to be a Data Scientist to Use Machine Learning?

The growth of augmented analytics and Machine Learning in manufacturing has made way for a new Data Scientist in the factory. With the increasing amount of connected devices and easy-to-use technology, more workers are enabled to leverage this data to make informed decisions. Still, this increase in analytics-driven decision making begs the question—do you need to be a Data Scientist to leverage the power of augmented analytics and Machine Learning?

The answer? No, and here’s why.

The Shift to Industry 4.0

The smart factory is no longer a thing of the future. For many manufacturers, it is the reality now. Data is a competitive advantage for manufacturers looking to increase productivity, sustainability, and overall OEE. Industry 4.0 has not only increased the amount of data accessible within the factory for analysis, but it has also increased the ability to analyze large, complex datasets, also known as Big Data. 

From AI-powered applications with predictive and prescriptive analytics, to easy-to-use data visualization software, the analysis and optimization of production processes is no longer solely the role of Data Scientists. Plant Managers have live production data at their fingertips to make real-time decisions. Process Engineers are able to analyze productivity based on historians and edge data whether they are on the shop floor or at corporate. With this shift, a new breed of analysts were created: the Citizen Data Scientist.

What is a Data Scientist vs. a Citizen Data Scientist?

The role of a Data Scientist is to model production data, which by necessity requires them to also build the processes and infrastructure for collecting this data based on an organization’s unique workflow and business objectives. A Data Scientist is an expert in data analysis and has a knack for solving complex data problems, but might be disconnected from the day-to-day production process. 

A Citizen Data Scientist is a newer role in manufacturing and a term coined by Gartner. Driven by the move to Industry 4.0, the role of the Citizen Data Scientist sprouted from the influx of more tracked and accessible data, and the use of predictive and prescriptive analytics. So, what is a Citizen Data Scientist? It is a person who conducts moderate to advanced diagnostic analysis by creating or generating models and analyzing that data. Where the two roles differ is that a Citizen Data Scientist’s primary job function (and typically, training) is outside the field of statistics and analytics.

A Citizen Data Scientist becomes a power user of advanced analytics tools over time, typically through on-the-job training and is enabled by easy-to-use technology. One key value of a Citizen Data Scientist is their ability to evaluate those augmented analytical tools with the context of what happens in the factory in mind. 

In manufacturing, a Citizen Data Scientist often works closely with Data Scientists to solve OEE challenges. The two roles are complementary to each other—where a Data Scientist typically holds the advanced data science expertise, a Citizen Data Scientist typically brings their own expertise and contextual knowledge.

Citizen Data Scientists in Manufacturing

The Citizen Data Scientist in a manufacturing setting typically holds an engineering role. Quality, process, maintenance, or chemical engineers are leveraging similar skill sets to Data Scientists to drive continuous improvement within their expertise. They may also hold a type of analyst or business management role.

Among other initiatives, Quality Engineers leverage Machine Learning apps to quickly and accurately identify defective goods and products early to reduce waste and uncover root cause. 

Process Engineers leverage Machine Learning apps to uncover hidden bottlenecks impacting OEE. With this discovery they are able to lean on their operational acumen and possibly a Digital Twin of the live production process to test and devise a solution.  

Maintenance Engineers are able to use Machine Learning apps to plan perfectly timed maintenance schedules (ie predictive or preventative maintenance, reducing unexpected (and costly) downtime.

It is possible to build many graphs with different ways to manipulate data. There is an algorithm to identify influential parameters very easily without being an expert of data analysis. There are different ways to survey data and tendencies with SPC cards or others graphics.”

— A Head of Data Analysis in the Manufacturing Industry

Enabling Data-Centric Decision Making

If data is the competitive advantage in manufacturing, organizations should start by building a data-centric decision making culture. Leveraging the power of Machine Learning to drive faster continuous improvement does not always require a Data Scientist. Enable all employees to innovate faster by providing them the tools and skills to analyze your production data through easy-to-use Machine Learning apps and IIoT platform.