#### Abstract

Milking machine vacuum regulators were tested at dairies across the United States over a period of twelve years. The drop in vacuum level with increasing air flow for each regulator tested was modeled using segmented regression. Three measures of regulator performance were considered: the slope of the line in the first phase, the variability about the first line, and the join point (after which vacuum pressure began to drop rapidly). The distribution of the join point was estimated based on an accelerated failure time model with censoring, Weibull errors, a model effect, and a linear effect of set point vacuum. For each model, the average slope of the first line, the average variability about the line of the first regime, and the estimated median cfm (cubic feet of air per minute, New Zealand standard) for a join point with set point vacuum of 13 in. Hg were standardized. These standardized values were used in a cluster analysis to identify four performance groups of regulator models.

#### Keywords

milking machine vacuum regulators

#### Creative Commons License

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

#### Recommended Citation

Young, Linda J.; Bodman, Gerald R.; Boilesen, Eugene C.; and Stroup, Walter W.
(1991).
"A STATISTICAL ANALYSIS OF THE PERFORMANCE OF MILKING SYSTEM VACUUM REGULATORS,"
*Annual Conference on Applied Statistics in Agriculture*.
http://newprairiepress.org/agstatconference/1991/proceedings/7

A STATISTICAL ANALYSIS OF THE PERFORMANCE OF MILKING SYSTEM VACUUM REGULATORS

Milking machine vacuum regulators were tested at dairies across the United States over a period of twelve years. The drop in vacuum level with increasing air flow for each regulator tested was modeled using segmented regression. Three measures of regulator performance were considered: the slope of the line in the first phase, the variability about the first line, and the join point (after which vacuum pressure began to drop rapidly). The distribution of the join point was estimated based on an accelerated failure time model with censoring, Weibull errors, a model effect, and a linear effect of set point vacuum. For each model, the average slope of the first line, the average variability about the line of the first regime, and the estimated median cfm (cubic feet of air per minute, New Zealand standard) for a join point with set point vacuum of 13 in. Hg were standardized. These standardized values were used in a cluster analysis to identify four performance groups of regulator models.