Authors
Kevin Potcner
JMP
Objective
Evaluate the performance to specifications of a food manufacturing process using confidence intervals and hypothesis testing.
Background
Constant monitoring of a manufacturing process is critical to ensure that a consistent product is produced and is able to meet the desired specifications. Food manufacturing is one such process where process quality is essential to ensure the consumer receives a consistent and safe product. Oakville Dairy is a food processing company that produces a variety of dairy-based products. A series of quality issues has recently surfaced in one of its yogurt products. In particular, some of the technicians in the testing lab have noticed rather excessive variation in the acidity levels (pH) between the batches of product. The Quality Engineering Team has been tasked with studying this through a more formal data collection and analysis process.
The Task
The facility runs eight separate production lines that make the yogurt. Ideally, each line should be producing product that is very similar between batches, as well as between the production lines. To get an overview of the current state of the process, the quality engineers take a sample of yogurt from each of the next 30 batches of product made by each of the eight production lines. The samples are taken to the lab and a variety of product characteristics, including Acidity (pH), is measured.
The specifications for Acidity (pH) level for the product is: target = 4.35; lower specification limit (LSL) = 4.30; upper specification limit (USL) = 4.40; and standard deviation <0.015.
They then perform inferential statistical analyses on each production line to determine if there is enough statistical evidence demonstrating that the process is not meeting specification.