* This blog was authored by Dr. Mohamed Abuali, CEO of IoTco, Predictronics’ newest business partner. IoTco and Predictronics have combined Predictronics’ award-winning PDX platform with IoTco’s best-in-class partner products, including MES and IIoT platforms, to offer innovative solutions for customers around the world wanting to digitally transform their global operations with predictive analytics technology. *

In the journey towards smart manufacturing through digitization, we often find ourselves mired in too many variables to analyze and too much data to handle. Big data gets bigger and soon we are swimming in our own data lake. Often there are self-doubts as to the direction we are going in and we tend to have second thoughts about the end results. It starts to feel like we are trying to boil the ocean, especially if the results are not promising.

It is crucial that we focus on the concrete steps to move from the beginning of a predictive analytics project to the end. Properly executing these key stages results in a successful deployment of predictive maintenance (worry-free uptime) and predictive quality (near-zero defects) solutions. These two areas are the low hanging fruit and the quickest wins for smart manufacturing today. The time has come where research in artificial intelligence and data analysis techniques are now becoming more readily available and ever-present as a packaged product within solution templates. These machine learning tools are practical and effective in countless industrial applications.

The six steps towards a successful implementation of a predictive analytics project in manufacturing are as follows:

1. Data Collection

What are the critical machine or sensor parameters for that application?

2. Data / Signal Processing

How do we clean and prepare the data for analysis?

3. Feature Extraction and Selection

What are the statistics derived from the raw data?

4. Health Assessment

What is your machine health?

5. Prediction

What is the remaining useful life of your machine?

6. Diagnostics

What is the root cause of failure?

Each step is critical to the overall success of the project and the positive business impacts created for the end user community. The true measure of success is the return on investment (ROI). Unless there is a direct connection to ROI, smart manufacturing will end up lost at the bottom of the ocean, much like artificial intelligence research once was. AI spent time waiting for CPU and RAM memory to play catch up to reach the stage we are at today. Now processing speeds can be utilized to make real-time computations of data to arrive at decisions in near real-time, enhancing AI value for real-world applications.

Combined with the ability to route data and decisions through the internet, the potential of predictive analytics and AI can now be realized by shop floor managers and the impact delivered to businesses across the globe.