Build a Predictive Analytics Model
Generate predictions and gain insight into your data with machine learning.
Guide Concept
This project will introduce ThingWorx Analytics Builder.
Following the steps in this guide, you will create an analytical model, and then refine it based on further information from the Analytics platform.
We will teach you how to determine whether or not a model is accurate and how you can optimize both your data inputs and the model itself.
Guide Info
| Level | Intermediate |
| Est. Time | 60 Minutes |
| Published | Jan 15, 2021 |
| Author | Jason Wyatt |
You'll learn how to
- Load an IoT dataset
- Generate machine learning predictions
- Evaluate the analytics output to gain insight
Things used in this guide
This guide may be completed in any environment in which you have access to both ThingWorx Foundation and Analytics.
This includes both the hosted evaluation server, as well as a combination of both the Foundation and Analytics Server downloadable trials.
ThingWorx AnalyticsBuild a Predictive Analytics Model
Challenges Manufacturers of factory equipment often rely on manual processes to maintain adequate supply levels of consumables in equipment at customer sites. But a clipboard-and-pen process for ordering supplies are error prone, and can quickly grow wrought with problems that often go unreported until they affect assembly line production, resulting in unexpected costs—such as overnight shipping charges—and risking downtime, which ultimately leads to customer dissatisfaction.
Solution One such manufacturer implemented ThingWorx to connect to equipment installed at customer sites, and created a custom web application using ThingWorx Foundation to remotely monitor supply levels. ThingWorx Industrial Connectivity provides the bi-directional connection to send data that is displayed in graphs on the web application created using ThingWorx Mashup Builder. Features of ThingWorx Analytics were used to generate alerts before maintenance problems affected production.
Outcomes The manufacturer is able to monitor supply levels to more effectively anticipate when consumables will need to be replenished. Supplies are now consistently ground shipped on time to meet assembly line demand, reducing interruptions to operations and allowing expansion to multiple plants with improved service level.