A network simulation model of large herringbone and parallel milking parlors

Area of Science:

• Agricultural Engineering
• Dairy Science
• Operations Research

Background:

• Efficient dairy farm management relies on optimizing milking parlor operations.
• Understanding the performance metrics of different milking parlor designs (herringbone vs. parallel) is crucial for maximizing productivity.

Purpose of the Study:

• To develop and validate a network simulation model for predicting the performance of herringbone and parallel milking parlors.
• To assess the accuracy of the simulation model in estimating cows milked per hour and milk harvested per shift.

Main Methods:

• A task-oriented network simulation model was constructed for herringbone and parallel milking parlors.
• Model components included personnel, vacuum pressure, pulsation ratio, and milk yield per cow.
• Probability distributions were fitted using data from four commercial dairies with varying parlor configurations.

Main Results:

• The simulation model accurately predicted cows milked per hour across double-16 and double-20 herringbone, and double-35 and double-40 parallel parlors.
• Simulated mean milk yields per shift closely matched observed yields for all parlor types.
• No significant differences were found between simulated and observed mean performance metrics.

Conclusions:

• The developed network simulation model provides a reliable tool for predicting milking parlor performance.
• The model can aid in optimizing dairy operations and evaluating the efficiency of different milking parlor designs.
• Accurate simulation of parlor throughput and milk yield is achievable with this modeling approach.