Adequate udder stimulation of dairy cows before milking is critical for the harvest of high-quality milk. It is essential for the milk-ejection reflex to obtain the alveolar milk, which represents approximately 80% of the udder’s milk volume. The intensity and duration of stimulation required for effective milk-ejection varies between cows, but generally increases at the end of lactation and with declining milk production. Traditionally, pre-milking stimulation has been achieved through milk stripping by hand. To date, dairy operations apply a fixed pre-milking stimulation regimen to all cows, irrespective of their physiological needs. This hampers our ability to elicit the maximum milk-ejection capacity particularly of late lactation cows resulting in delayed milk ejection (DME) in approx. 25% of cows on New York State dairies. Delayed milk ejection leads to poor milking efficiency, impaired teat and udder health, and reduced milk yield. Conservative estimates suggest a loss in milk production of 7.5 kg/day in cows with DME. Given that 25% of New York State´s dairy cows suffer daily from DME, we estimate that DME results in an income loss of approx. $250,000/year on a 1,000-cow dairy.

With the advancement of milking technology, alternative forms of pre-milking stimulation through automated systems have been developed. However, the potential of these automated pre-milking stimulation (APS) systems for providing supplemental stimulation to cows with DME has not been investigated. We lack objective and automated methods to efficiently identify cows with DME in need of an enhanced stimulation protocol, and to assign them in an automated manner. In this study, we propose to develop a method to automatically identify cows with DME and maximize their milk production through application of supplemental stimulation using an APS system. First, we will use milk flow data from electronic on-farm milk meters to identify cows with DME in a prospective case-control study. Differences in milk yield between cases and controls will be assessed and used to determine the most discriminatory case definition for DME. Second, we will validate the effect of supplemental pre-milking stimulation on milk production of DME cows using an existing APS system in a randomized controlled trial. Last, we will extend an existing infrastructure for integration of data from electronic on-farm milk meters and develop an algorithm that will generate a list of cows at risk of DME for which immediate actionable decisions can be made according to the farm-goals.

New York State, ranking third in national milk production, has dairy cash receipts over $3 billion dollars annually. The dairy industry is thus vital to the economic well-being, sustainability, and growth of our communities. Accommodating the physiological requirements of individual cows in a precision dairy farming system is of utmost importance to optimal milk harvest and animal well-being. To achieve this, a new dimension for automated identification of cows with DME and providing them with additional pre-milking stimulation using APS systems is critical. Our interdisciplinary team of investigators is experienced in the proposed methodologies and highly integrated into the New York State dairy industry. This will enable a seamless transfer of the knowledge gained from this work to immediate application in the field. By developing automated monitoring systems to identify cows with DME, then managing them according to their individual needs, dairy herds will be more productive. We anticipate that our project will generate opportunities for cross campus collaborations. Further, our long-term goal to effectively utilize and integrate new technologies to enhance U.S. food and agriculture enterprises aligns well with the USDA-NIFA-AFRI Agriculture Systems and Technology (AST) program area priority “Engineering for Agricultural Production Systems” (A1521). Data from this proposed CIDA project will uniquely position us for a successful grant proposal.