What is Statistical Process Control (SPC)?
Statistical Process Control (SPC) is a method of quality control which uses statistical methods to monitor and control a process. Its main goal is to ensure that a process operates efficiently, producing more specification-conforming products with reduced waste.
SPC is applied in situations where it's important to keep a process stable and predictable, helping to prevent defects rather than detect them after they've occurred. By analyzing data from a process, SPC helps identify and eliminate sources of variation, leading to improved quality and efficiency.
What are Control Charts?
Control charts are a fundamental tool of SPC. They are graphs used to study how a process changes over time. Data are plotted in time order. A control chart has a central line for the average, an upper control limit (UCL), and a lower control limit (LCL).
These limits are determined from historical data and indicate the range within which the process is expected to operate when it is "in control." Points outside these limits, or specific patterns within the limits, suggest that the process is affected by "special cause variation" and needs investigation.
Types of Control Charts and When to Use Them
Control charts are broadly categorized into two types: charts for variables data and charts for attributes data.
Control Charts for Variables Data
Variables data are measurements that can be quantified on a continuous scale (e.g., length, weight, temperature, time). These charts are used when you can measure the characteristic of interest.
X-bar and R Chart
- What it is: Used for processes that have subgroup sizes between 2 and 10. The X-bar chart monitors the process mean, while the R chart monitors the process range (variability).
- When to use: Ideal for monitoring the central tendency and spread of a process when data can be collected in small subgroups. Common in manufacturing for characteristics like part dimensions.
X-bar and S Chart
- What it is: Similar to the X-bar and R chart, but used when subgroup sizes are larger (typically 10 or more). The X-bar chart monitors the mean, and the S chart monitors the standard deviation (a more robust measure of variability for larger subgroups).
- When to use: Preferred over X-bar and R charts for larger subgroup sizes as the standard deviation provides a better estimate of process variation.
Individuals and Moving Range (I-MR or X-MR) Chart
- What it is: Used for processes where it's difficult or impossible to collect data in subgroups (i.e., subgroup size is 1). The Individuals chart monitors individual observations, and the Moving Range chart monitors the variation between consecutive individual observations.
- When to use: Suitable for slow processes, automated processes where each item is measured, or when there's a long time between data points. Examples include monitoring monthly sales, daily temperature, or individual chemical batch yields.
Control Charts for Attributes Data
Attributes data are counts of non-conformities (defects) or non-conforming items (defectives). These charts are used when you can only count, not measure, the characteristic of interest.
p-Chart (Proportion of Defectives)
- What it is: Monitors the proportion of non-conforming items in a sample. The sample size can vary.
- When to use: Used when you want to track the percentage of defective items (e.g., percentage of faulty products in a batch, percentage of late deliveries).
np-Chart (Number of Defectives)
- What it is: Monitors the actual number of non-conforming items in a sample. Requires a constant sample size.
- When to use: Similar to the p-chart, but used when you prefer to work with the actual count of defectives rather than their proportion, and your sample size is constant.
c-Chart (Number of Non-conformities)
- What it is: Monitors the number of non-conformities (defects) per unit when the number of inspection units is constant. A unit can have multiple non-conformities.
- When to use: Used when counting defects within a fixed unit (e.g., number of scratches on a car panel, number of errors per page of text, number of calls per hour in a call center).
u-Chart (Number of Non-conformities per Unit)
- What it is: Monitors the number of non-conformities (defects) per unit when the number of inspection units varies.
- When to use: Similar to the c-chart, but used when the "opportunity for defects" (i.e., the size of the inspection unit) varies from sample to sample (e.g., number of errors per software module where module size varies, number of accidents per flight where flight duration varies).
