Generalized Additive Mixed Modeling of Total Transport Losses of Market-Weight Pigs

The objective of this study was to investigate factors associated with total transport losses (TTL) of market-weight pigs using a Generalized Additive Mixed Model (GAMM). The dataset was provided by Iowa Select Farms (Iowa Falls, IA), and included the information on 26,819 shipments delivered to 2 abattoirs. TTL was fitted as an overdispersed binomial process using a GAMM which included the fixed effects of abattoir, type of driver (i.e. truck owner or employee), distance traveled, average market-weight, wind speed, precipitation, and temperature-humidity index (THI), as well as the random effects of truck company and the combination of farms and quarter of the year. The estimated risk of losses was 0.867 (95% CI: 0.865-0.870) times lower when trucks were driven by owners instead of by employees, suggesting that truck owners have more vested interest. THI was associated with TTL (P < 0.0001, Figure 1A), displaying greater risk of losses at its extremes, indicating that additional care must be considered at these levels. The interaction between wind speed and precipitation was associated with TTL (P = 0.0209, Figure 1B), indicating a complex relationship between both explanatory variables and TTL. The distance traveled was associated with TTL (P = 0.0034, Figure 1C), with increased losses at distances up to 125 km and decreased risk of losses afterwards. This result suggests that long trips may give extra time to pigs to recover from the prior stress incurred at loading. The interaction between average market weight and abattoir was positively associated with TTL (P < 0.0001, Figure 1D), indicating a faster increment in the risk of losses in one facility relative to the other. In conclusion, TTL of market-weight pigs are caused by a complex system involving multiple interacting factors. Furthermore, GAMM is shown to be a flexible predictive tool capable of modeling non-linear relationships and which can be used to support decision-making in swine industry.

Guilherme Rosa
Guilherme Rosa
University of Wisconsin-Madison