Burks T F, Turner L W, Crist W L
Biosystems and Agricultural Engineering Department, University of Kentucky, Lexington 40546, USA.
J Dairy Sci. 2006 Jun;89(6):2343-52. doi: 10.3168/jds.S0022-0302(06)72305-0.
A time and motion study was conducted at 13 small dairy farms with average herd sizes less than 100 cows. Parlors were configured with 3 to 6 stalls per side. A data acquisition methodology was developed using a video camera to gather work routine time data in the parlors. A computer-based data logger was used to extract individual event durations during video playback. Each parlor's video record was reviewed in the laboratory so that work routine times across all parlors and operators could be pooled to estimate typical operator performance. There were 34 operator work routine times associated with various procedures in milking parlors that were evaluated in this study. Individual times were compiled for each work routine and a data-fitting program called UNIFIT was used to fit the data to 1 of 4 data models: gamma, lognormal, Weibull, and Pearson #5. Each of 34 work routine variables was fitted, tested, and plotted to determine how well each of those models fit the actual data. Distributions for Pearson #5, lognormal, gamma, and Weibull models were best fitted to 12, 10, 8, and 4 work routine times, respectively. More common tasks such as attaching the milker, grabbing a towel, and drying the udder were more consistently executed and had smaller variances than routines in which the operator would leave the pit to go to the milk room or disassembled the milk collector after milking. One of the better fitting models was the lognormal distribution for the time to "attach milker," which had a low relative discrepancy to the P-P plot (model probability vs. data probability) of 0.019 and a moderate chi(2) test value of 0.358, thus demonstrating a good fit of the model to the data. Simulation tests were compared with observed data to validate models for work routine times and demonstrated that the models accurately predict parlor throughput in small- to medium-sized parlors.
在13个平均牛群规模小于100头奶牛的小型奶牛场进行了一项时间与动作研究。挤奶厅每侧配置有3至6个挤奶位。开发了一种数据采集方法，使用摄像机收集挤奶厅的工作流程时间数据。使用基于计算机的数据记录器在视频回放期间提取各个事件的持续时间。在实验室对每个挤奶厅的视频记录进行审查，以便汇总所有挤奶厅和操作员的工作流程时间，以估计典型的操作员绩效。本研究评估了与挤奶厅各种程序相关的34个操作员工作流程时间。为每个工作流程编制了单独的时间，并使用一个名为UNIFIT的数据拟合程序将数据拟合到4种数据模型中的一种：伽马分布、对数正态分布、威布尔分布和皮尔逊5型分布。对34个工作流程变量中的每一个进行拟合、测试和绘图，以确定这些模型中每一个与实际数据的拟合程度。皮尔逊5型分布、对数正态分布、伽马分布和威布尔分布模型分别最适合12个、10个、8个和4个工作流程时间。与操作员离开挤奶坑前往牛奶室或挤奶后拆卸集奶器的流程相比，诸如连接挤奶器、拿毛巾和擦干乳房等更常见的任务执行得更一致，且方差更小。“连接挤奶器”时间的较好拟合模型之一是对数正态分布，其与P-P图（模型概率与数据概率）的相对差异较低，为0.019，卡方检验值适中，为0.358，从而表明该模型与数据拟合良好。将模拟测试与观测数据进行比较，以验证工作流程时间模型，并证明这些模型能够准确预测中小型挤奶厅的产量。
