Nowadays, ensuring product quality should be one of the key elements of any business. To achieve this goal, companies utilize various tools, including Statistical Process Control (SPC).
SPC enables the monitoring of production processes, allowing for the early identification of errors and faulty processes. This early detection enables timely intervention through appropriate corrections, modifications, or improvements to maintain quality standards.
Basic Concepts Used in SPC
Our understanding of SPC begins with an exploration of fundamental statistical concepts. This foundational knowledge will provide a framework for the concepts discussed later on.
σ – (Sigma)
Sigma, represented by the 18th letter of the Greek alphabet, signifies Standard Deviation in mathematics. Standard deviation is a measure of variability, indicating the distribution or spread around the mean value of any process. It’s important to note that higher standard deviations correspond to greater variability in characteristic values within the process.
Fig. 1 Graphical interpretation of the Gaussian distribution
Mode (Dominant)
The mode represents the most frequent value in a dataset.
Median
The median is the middle value of a dataset, where half of the observations are above it and half are below it. In a perfectly symmetrical distribution, such as the normal distribution, the mean, mode, and median are equal.
Points of Inflection
These are the points on a curve where the direction of concavity changes. In a normal density curve, the inflection points occur at one standard deviation away from the mean.
Fig. 2: Graphical Interpretation of Inflection Points
Control by Detection vs. SPC
Similar to problem-solving, where we can reduce Occurrence or increase Detection, the same principle applies to process control management.
Detection
The main reasons why control by detection is ineffective in terms of time and cost are:
– Control (especially 100%) of products can be time-consuming, costly, or impossible.
– “Process control” is limited to corrections in case of detected non-conformities.
– Depending on the process and the established “trigger point” for process adjustment, the process may exhibit significant inertia.
Fig. 3: Graphical Interpretation of Process Control by Detection
SPC – Statistical Process Control
Unlike detection-based control, statistical process control (SPC) relies on assessing a representative sample. Additionally, SPC involves reacting to signals indicating process degradation, enabling earlier intervention. Most importantly, it enables intervention before a non-conforming product is produced.
Fig. 4: Graphical Interpretation of Statistical Process Control
Reasons for Production Process Changes
The production process can undergo changes due to various factors, including time, operator influence, tool wear, environmental conditions, and material characteristics. These factors are among the primary influences contributing to process variations.
Fig. 5. Factors Influencing Changes in the Production Process
SPC – Basic Production Capability Indicators
When developing SPC, we encounter two main indicators, detailed below:
Potential Capability:
– measures the capability that a process can achieve under perfect centering.
– disregards the position of the distribution relative to the specified data.
– depends on the process spread, defined as the range where 99.73% of data falls (for a normal distribution, ±3σ).
– depends on the width of the tolerance interval (USL-LSL), where USL represents the Upper Specification Limit and LSL represents the Lower Specification Limit.
– compares the distribution’s “size” to the specified tolerance width.
– its value cannot be less than or equal to 0.
Actual Capability
Actual Capability assesses a process’s capability by comparing the process spread width to the tolerance width, considering the process’s position relative to specification limits.
It depends on both the position and the spread.
Its value cannot exceed that of Cp, Pp, or Cm
Sources of Variability in SPC
Two concepts related to the sources of variability are inherently linked to the concept of SPC: common and special.
Common Sources of Variability
Numerous factors contribute to common variability, each exerting relatively minor and random effects. These random causes are typically insignificant and do not warrant intervention in the process. Common sources of variability include:
Random Factors: Unpredictable events affecting the production process, such as machine breakdowns, human errors, or changes in weather conditions.
Systematic Factors: Regularly occurring influences that consistently impact the process, such as inadequate tolerances, incorrect process parameters, or inappropriate tools.
Variability in Materials: Differences in properties among raw materials or subcomponents used in the production process, which can affect product quality.
Variability in Operator Performance: Differences in the skills, experience, and motivation of operators involved in the production process.
Environmental Conditions: Workplace conditions such as temperature, humidity, pressure, lighting, or noise that can influence the production process and product quality.
Delivery Errors: Mistakes in the delivery of materials, parts, or finished products.
Special Sources of Variability
These have a non-random nature. The occurrence of a special cause disrupts the process and often requires intervention.
Please also remember that common causes should be minimized, while special causes should be eliminated.
SPC – What are the benefits of proper implementation
Statistical Process Control is one of the key elements affecting the main indicators of a company. With properly implemented SPC, we have the opportunity to:
Improve product quality by continuously monitoring production and service processes and eliminating errors in real-time.
Optimize processes and reduce costs by identifying sources of errors and faulty processes that affect product and service quality.
Increase employee engagement by involving them in the quality control process and engaging them in process improvement.
Enhance the company’s reputation by offering high-quality products, which contributes to customer satisfaction and builds trust.
Provide accurate data and information that assist in making informed decisions and achieving company goals.
Dariusz Kowalczyk
