Sorry, no posts matched your criteria.
Imagine a beverage factory producing thousands of bottles every day with a target volume of 600 mL per bottle, even though the machine has been calibrated, there will always be slight variations in the volume produced due to the natural characteristics of the production process.
The question is, when can such variations still be considered normal and when should they be a sign of a problem in the process? In Statistical Process Control (SPC), these conditions are analyzed using the Upper Control Limit (UCL) and Lower Control Limit (LCL), which are the upper and lower control limits used to evaluate whether a process is still in statistical control. Unlike specification limits, which are determined based on customer requirements, the UCL and LCL are calculated based on the natural variation of the process, thus serving as early indicators of the emergence of assignable causes of variation (Montgomery, 2020).
In a control chart, the UCL and LCL serve as control limits for observed quality characteristics, such as product dimensions, weight, temperature, and process time. As long as the observation points fall between these two limits and do not show any unusual patterns, the process is considered to be in statistical control. Conversely, if a point exceeds the UCL or falls below the LCL, or if a specific pattern emerges such as a continuous upward trend, a consecutive decrease, or a recurring cycle, then the process is suspected to be influenced by a special cause that requires further investigation (Besterfield, 2018). Thus, control charts are not only used to detect defective products, but also as a preventative tool to identify process deviations before they result in greater losses.
Determination of the UCL and LCL values is carried out through a statistical approach based on historical process data. In general, the center line represents the average value of the process, while the control limits are calculated using the principle of three standard deviations (±3σ) from the mean. For the X̄ chart, for example, the UCL and LCL values are obtained by utilizing certain constants adjusted for sample size, while in the Individuals Chart the control limits are calculated directly using the estimated process standard deviation (Montgomery, 2020). The ±3σ approach was chosen because it statistically covers approximately 99.73% of the natural variation of the process, which follows a normal distribution. Therefore, data outside the control limits has a very small chance of occurring naturally and is an indication of a change in process conditions that requires immediate analysis.
The application of UCL and LCL can be found in various manufacturing sectors. For example, in the machining process of shafts, the product diameter is periodically measured and plotted on a control chart.
If all measurement results are between UCL and LCL, then the process is considered stable even though there are still small variations between products. However, if some observation points suddenly exceed the UCL, the condition may indicate tool wear, machine calibration errors, or changes in cutting parameters. By knowing these causes early on, operators can immediately make adjustments before producing large quantities of product that is outside specifications. A similar approach is also applied in the food, pharmaceutical, electronics and automotive industries to maintain consistent quality and reduce costs due to scrap and rework (Evans and Lindsay, 2020).
Thus, the Upper Control Limit (UCL) and Lower Control Limit (LCL) are crucial parameters in statistical-based quality control. These two control limits enable companies to distinguish between natural process variations and those caused by specific disturbances, allowing timely corrective action. Through the implementation of SPC supported by UCL and LCL, organizations are able to maintain process stability, improve product quality consistency, reduce waste, and strengthen competitiveness amidst the demands of modern industry that increasingly prioritize quality and operational efficiency.
Writer: Brian Arga Prasidio Putra
Editor: Brian Arga Prasidio Putra
Reference
Besterfield, D.H. (2018). Quality Control. Edisi ke-9. Boston: Pearson.
Evans, J.R. dan Lindsay, W.M. (2020). Managing for Quality and Performance Excellence. Edisi ke-11. Boston: Cengage Learning.
Montgomery, D.C. (2020). Introduction to Statistical Quality Control. Edisi ke-8. Hoboken, NJ: Wiley.
Shewhart, W.A. (1931). Economic Control of Quality of Manufactured Product. New York: D. Van Nostrand Company.
Check out the full news on our social media:
Website: https://www.its.ac.id/tindustri/laboratorium-sistem-manufaktur/ Instagram: @sismanity LinkedIn: www.linkedin.com/in/manufacturing-systems-laboratory YouTube: SISMANITY ITS
