scale stabilization, weight precision, efficiency
This trial was conducted to evaluate the optimum scale head settings for LeeO scale systems to balance accuracy, precision, and efficiency in weighing pigs. LeeO scales allow for the adjustment of both stabilization time (mSec) and stabilization weight (g). The pig weight that is accepted by the technology is only registered after staying within the set stabilization weight for the set stabilization time. Prior to beginning animal research, precision and accuracy were assessed for two scales. The nursery scale used a 25-lb test weight and the finisher scale used a 50-lb test weight. The CV estimates were 0.1% or less when the test weight was measured multiple times within each combination of settings for both nursery and finishing weights. Accuracy did not differ (P > 0.10) based on stabilization time for either scale. However, when weighing the 25-lb test weight, the longest stabilization time of 1,000 mSec resulted in the smallest difference from the true weight (P < 0.05). To assess scale settings, 30 nursery and 33 finishing pigs were weighed multiple times using different settings to determine accuracy, precision, and efficiency. Each pig was weighed 5 times on the predetermined settings for a total of 45 weights for each nursery pig and 20 weights for each finishing pig. Coefficient of variation (CV) was used as an estimation of precision, which was calculated by dividing the standard deviation of the 5 weights for that combination of scale head settings by the average weight of the 5 weights for that setting. To estimate accuracy, the absolute difference of the average weight of the 5 weights for that setting combination from the overall average weight for that pig was calculated. Efficiency was measured one of two ways. The nursery pig procedure included the elapsed time from when the first weight was collected for that combination of scale head settings until the fifth weight that was collected was divided by the total number of weighing events for that setting. The finishing pig procedure included the sum of the times that it took to lock in all five weights for that combination of scale head settings and dividing that sum by the total number of weighing events for that setting. There were no differences in accuracy for nursery or finishing pig scales based on stabilization time or weight (P > 0.10). There was a significant difference in precision for the nursery pig scale based on both stabilization time (P = 0.003) and stabilization weight (P = 0.003); with CV improving as stabilization weight became smaller and stabilization time became longer. Conversely, efficiency for collecting both nursery and finishing weights improved with larger stabilization weight and shorter stabilization time (P < 0.001). In the finishing experiment, CV was improved (P < 0.05) for the 500 g × 1,000 mSec and 500 g × 500 mSec settings compared to the 1,000 g × 250 mSec setting, with the 1,000 g × 500 mSec setting intermediate. To balance precision and efficiency, a setting of 50 g × 500 mSec for the nursery and 500 g × 1,000 mSec for the finisher is recommended.
Jenkins, Abigail K.; Rao, Zhong-Xing; Gebhardt, Jordan T.; Woodworth, Jason C.; Tokach, Mike D.; DeRouchey, Joel M.; Goodband, Robert D.; Holck, Tyler; Stave, Joel; and Christensen, Claire
"Evaluation of Precision, Accuracy, and Efficiency of Scale Stabilization Settings Using LeeO Pig Tracking System,"
Kansas Agricultural Experiment Station Research Reports: