Statistical Process Control Template for Spotfire® using Spotfire® Data Science - Workbench 2.1.0

Overview

The basic building block of this template is the quality control chart. A control chart is used to monitor a process variable over time. The objective of monitoring is to determine if the process is in control or out of control. This helps to ensure that the process is stable over time and minimizes its output variability.  It can be applied in different fields like manufacturing, telco, energy, marketing, sales, healthcare, etc.

This template provides various quality control charts and tests relevant to process control:  you can construct charts for monitoring the process mean as well as variance, and you can create charts for individuals or X-bar charts. Here is the list of charts: X type charts for individuals, X-bar, EWMA, Moving average, Cusum, R/S charts, and histograms.  They can include central lines, control, warning as well as specification limits.

Points outside of limits are not the only way to identify atypical process behavior. Other tests implemented in this template are Runs Tests which identify a set of consecutive points that are not behaving randomly (e.g. 6 consecutive points with increasing or decreasing trend). All Rules Tests are configurable from inside the dashboard.

Implementation  

This component consists of two applications.

• The first one - Application 1 - is an interactive dashboard application with extensive manual options to configure charts.  The user can interactively change almost every setting using document properties and parameters. Additional capabilities implemented in this application are Brushing (recalculating charts with excluded samples) and assigning causes, actions, and comments. This application is a good place to start defining and experimenting with various chart types, rules, and configurations.
• If the user wants to investigate in an automated way a larger number of parameters with predefined configurations for quality control chart types and alarms, Application 2 can be used. All the definitions for each parameter are driven by one configuration line inside a configuration file. To run quality control charts for your data, all you need is to upload your configuration file and the measurement data. 

This template is working at the moment for the English locale.

We recommend reviewing a detailed article Process Control Monitoring and Alerting Solution describing the full solution which utilizes and extends capabilities of this template.