Search
Author
Title
Vol.
Issue
Year
1st Page

# Journal of Animal Science - Animal Products Predicting pork loin chop yield using carcass and loin characteristics

1. Vol. 94 No. 11, p. 4903-4910

Received: May 06, 2016
Accepted: Aug 17, 2016
Published: October 27, 2016

1 Corresponding author(s):
View
Share

doi:10.2527/jas.2016-0610
1. K. B. Wilson*,
2. M. F. Overholt*,
3. E. K. Hogan*,
4. C. Schwab,
5. C. M. Shull,
6. M. Ellis*,
7. N. S. Grohmann*,
8. A. C. Dilger* and
9. D. D. Boler 1*
1. * Department of Animal Sciences, University of Illinois, Urbana-Champaign 61801
The Maschhoffs, Carlyle, IL 62231

## Abstract

The objective was to determine the predictive ability of carcass length for the number of equal-thickness chops obtained from a boneless pork loin. Longer pork carcasses are assumed to yield longer loins and, therefore, an increased number of chops. Loins were collected from pigs (1,238 total) raised under commercial conditions and marketed when the mean pig weight in a pen reached 138 kg. Pigs were slaughtered over 7 wk in a commercial facility. Carcass length was measured at 1 d postmortem on the left side of each carcass from the anterior edge of the symphysis pubis bone to the anterior edge of the first rib. Carcasses were fabricated, and boneless loins (North American Meat Processors number 414) were vacuum packaged and transported to the University of Illinois Meat Science Laboratory. Loins were stored at 4°C for 14 d. At the end of the aging period, loins were weighed, measured for stretched length (stretched to maximum length without distortion) and compressed length (compressed to minimum length without distortion), and sliced into 2.54-cm-thick chops. Boneless chops were counted and weighed. Carcass length ranged from a minimum of 78.2 cm to a maximum of 96.5 cm and the number of boneless chops ranged from a minimum of 13 to a maximum of 20 chops. Data were analyzed using the regression procedure of SAS. The dependent variable was the number of boneless chops. Coefficient of determination (R2) was calculated for carcass length, boneless loin weight, compressed loin length, and stretched loin length. Carcass length explained 15% (P < 0.0001) of the variation in the number of loin chops. Loin weight explained 33% (P < 0.0001) of the variation in the number of loin chops. Compressed loin length and stretched loin length explained 28 and 8% (P < 0.0001), respectively, of the variation in the number of loin chops. Multiple linear regression was used to determine a predictive equation for the number of loin chops using the stepwise selection option of all independent variables. The combination of boneless loin weight, compressed loin length, 10th-rib carcass fat depth, and carcass length explained 45% of the variation (P < 0.0001; C(p) = 16.76) in the number of loin chops using a required F statistic at the SLENTRY and SLSTAY level = 0.15. Overall, carcass length is a poor predictor of the number of equal-thickness loin chops that can be derived from a boneless pork loin.

## INTRODUCTION

Pork carcass length is heritable (estimate = 0.62; Lo et al., 1992) and linearly increased with heavier carcasses (Cisneros et al., 1995). Furthermore, carcass length was correlated (r = 0.49, P < 0.01) with estimated lean percentage of pigs (Trew et al., 1987). However, removing the backbone increased error associated with measuring carcass length of all carcasses (Braude et al., 1957). Recently, the retail price of boneless pork chops increased approximately 17% from $7.92/kg in 2006 to$9.56/kg in 2015 (Bureau of Labor Statistics, 2015). At the same time, the price of carcasses increased only 6% (Schulz, 2016). Furthermore, 30% of pork is consumed as bone-in or boneless pork chops (USDA, 2005). Therefore, if longer carcasses produce more boneless loin chops, it stands to reason that longer carcasses would generate more revenue. However, without the backbone and after the resolution of rigor mortis, a boneless intact pork loin can be distorted and, therefore, carcass length may not truly reflect loin length. Still, measuring boneless loin parameters such as compressed and stretched loin length may be more predictive of the number of boneless chops. Given that carcass length was actually based on the fixed length of the skeleton and not the variable length of muscle, the hypothesis was that carcass length would not be predictive of loin chops obtained from a boneless loin but other carcass or loin characteristics may be.

If boneless loin parameters such as boneless loin weight, stretched length, and compressed length are predictive of the number of boneless chops, they may be valuable to packers. Even so, predictive ability estimates of carcass length, boneless loin weight, stretched loin length, and compressed loin length on the number of loin chops have not been determined. Therefore, the objective was to use linear regression to determine which carcass and boneless loin traits were the most predictive of the number of loin chops derived from a boneless loin.

## MATERIALS AND METHODS

Pigs were slaughtered under the inspection of the USDA Food Safety Inspection Service at a federally inspected facility. Boneless loins were purchased from that facility and transported to the University of Illinois Meat Science Laboratory (Urbana, IL). Therefore, Institutional Animal Care and Use Committee approval was not obtained.

### Experimental Design and Processing Facility Data Collection

Loins were obtained from both barrows and gilts (1,238 total) from a single genetic line and were raised and slaughtered under commercial conditions. Pigs were housed in single-sex pens with 20 pigs per pen. Five pigs from each pen were selected for evaluation. The 5 selected pigs represented a pig with an ending BW closest to the pen average BW, a pig with a BW 1 SD above and a pig with a BW 1 SD below the average BW of the pen, and a pig with a BW 2 SD above and a pig with a BW 2 SD below the average BW of the pen. Pigs were slaughtered over 7 wk as the average BW of each pen reached 138 kg. Transportation distance was approximately 277 km and pigs were held overnight with no access to food but free access to water prior to slaughter. Pigs were immobilized via carbon dioxide stunning and terminated via exsanguination. Following commercial slaughtering procedures, HCW was measured along with fat depth and loin depth using a Fat-O-Meater probe (SFK Technology A/S, Herlev, Denmark) at approximately the 10th rib location and estimated carcass lean was calculated using the facility’s proprietary equation. Carcasses were blast chilled for approximately 90 min. After exiting blast chill, carcasses with minimal harvest trim and complete carcass characteristic estimates were identified and placed in a temperature equilibration cooler. After an equilibration period of approximately 1 h, carcasses were identified with a slaughter sequence on the vertebral column of the loin. Carcass length was measured on the left side of each carcass from the anterior edge of the symphysis pubis bone to the anterior edge of the first rib. Carcasses were fabricated at approximately 22 h postmortem into primal pieces. Loins were separated from the shoulder between the second and third ribs and separated from the ham 2.79 to 3.81 cm anterior to symphysis pubis bone. Loins were further fabricated into boneless Canadian back loins (NAMP number 414; NAMP, 2007), vacuum packaged, and transported to the University of Illinois Meat Science Laboratory for further evaluation.

### Boneless Loin Chop Determination

Loins were aged for 14 d at 4°C, after which they were removed from their packaging and weighed to determine boneless loin weight. Stretched loin length (stretched to maximum length without distortion; lack of distortion was subjectively determined) and compressed loin length (compressed to minimum length without distortion) was measured by hand using a tape measure (Prym Consumer USA, Spartanburg, SC) on each loin prior to slicing to the nearest 0.5 cm. Then, loins were sliced into 2.54-cm-thick chops using a push-feed style Treif Puma slicer (Treif model 700 F; Treif, Oberlahr, Germany). Ends and incomplete chops (chops from the blade end that were distorted during slicing and any chops not 2.54 cm thick) were assessed and weighed. Complete 2.54-cm-thick boneless chops were counted. Chop yield was calculated: [(boneless loin weight, kg − ends and pieces, kg)/boneless loin weight, kg] × 100. Total chop weight per loin was calculated as boneless loin weight, kg − ends and pieces weight, kg. Total chop weight per loin was used to calculate the breakeven price of each loin. Breakeven price was calculated as [boneless loin weight, kg × boneless center cut loin, strap-off price ($1.34/lb, converted to$2.95/kg; USDA AMS, 2015)] − [ends and pieces weight, kg × 72 trim, combo price ($0.82/lb, converted to$1.81/kg; USDA AMS, 2015)]. Revenue from total chops per loin was calculated as total chop weight per loin, kg × $9.66/kg (convert1ed from$4.39/lb; Bureau of Labor Statistics, 2015).

### Statistical Analyses

Because carcass length was a function of the individual pig, loin served as the experimental unit for each set of analyses. Population summary statistics were calculated with the MEANS procedure of SAS (version 9.3; SAS Inst. Inc., Cary, NC). Pearson correlation coefficients among independent variables were calculated using the CORR procedure of SAS. Correlations were considered significant at P < 0.001. Other data were analyzed using the REG procedure in SAS. The dependent variables were the number of chops produced from a boneless loin or chop yield. Coefficients of determination (R2) were calculated for following independent variables: carcass length, HCW, compressed loin length, stretched loin length, loin depth, and boneless loin weight. A linear regression equation was developed using the independent candidate variables to predict the number of loin chops derived from a boneless loin. An initial regression model included each of the 6 independent variables as well as 10th-rib carcass fat depth. Multicollinearity among independent variables was assessed using a variance inflation factor (VIF) statistic. However, no parameters exceeded VIF values of 4; therefore, all independent variables remained as candidate variables for selection in the model. Influence of individual observations on the estimated dependent variable was determined using the difference of fit (DFFITS) statistic. Observations were considered to have excessive influence on the estimation of the regression parameters when DFFITS ≥ 2[(p/n)1/2], in which p = was the number of parameters considered and n is the total number of observations. In the present study, 7 variables were considered and 1,238 observations were used. Twenty observations met this criterion and were removed from the data set. Using the stepwise selection method, independent variables were required to have a significant F statistic at the SLENTRY and SLSTAY level = 0.15 to be included and remain in the final model.

## RESULTS AND DISCUSSION

### Carcass Characteristics

Population summary statistics including mean, minimum observation, maximum observation, and CV are presented in Table 1. The mean HCW of pigs from this trial was 103.6 kg. The selection strategy implemented resulted in wide variation in HCW, as it ranged from 75.0 kg to 131.0 kg with a CV of 9.76. Moreover, neither fat thickness nor LM depth was controlled; therefore, the selection criteria resulted in calculated CV estimates of 22.40 for back fat (BF) depth and 10.84 for LM depth. However, carcass length (CV = 3.64) was relatively less variable than other carcass compositions traits. Other studies that slaughtered pigs over wide ranges in ending BW reported similar CV estimates. Carcasses with an average length of 77.44 ± had the same chilled carcass weight CV (12.22) and length CV (4.47) regardless of fat thickness category ((2.29-3.28 cm, 3.30-4.29 cm and 4.32-5.33 cm, Edwards et al., 1981). Likewise, CV estimates were 4.45 for carcass length and 12.67 for chilled side weight in a population of pigs that ranged in chilled side weight of 32.78 ± 4.15 kg (Cross et al., 1975). Differences in CV estimates between BW and carcass weight can be easily explained because carcass length is indicative of bone growth and reflects the overall size of the skeleton. Furthermore, pigs attain a large proportion of skeletal size before reaching mature BW (Gerrard and Grant, 2003). Because of this, measurements that have evaluated skeletal growth (carcass length) as a proportion of BW decrease as the animal gets heavier (Gerrard and Grant, 2003). Over 40 yr ago, chilled carcass weights were approximately 68.3 kg and carcass length was approximately 76.5 cm (Pearson et al., 1970). Carcasses in this study averaged 103.6 kg and were 86.8 cm long (Table 1). This demonstrates a 41.1% increase in HCW but only a 12.6% increase in carcass length.

View Full Table | Close Full ViewTable 1.

Population summary statistics of carcass and loin characteristics

 Item No. Mean Minimum Maximum CV, % HCW, kg 1,235 103.6 75.0 131.0 9.76 Fat depth, mm 1,234 22 11 51 22.40 Loin depth, mm 1,234 67 36 86 10.84 Boneless loin weight, kg 1,232 4.1 2.4 6.1 14.95 Carcass length, cm 1,238 86.8 78.2 96.5 3.64 Stretched loin length, cm 1,235 63.5 53.0 74.0 4.86 Compressed loin length, cm 1,237 51.5 42.5 62.5 6.16 End and pieces weight, kg 1,235 0.43 0.19 0.91 24.20 Chops, no. 1,238 17 13 20 7.82 Chop yield, %1 1,236 89.5 81.8 96.0 2.64
1Chop yield was calculated as [(boneless loin weight, kg − ends and pieces, kg)/boneless loin weight, kg] × 100.

Even so, in this population of carcasses, carcass length was correlated (r = 0.71) with HCW (Table 2). This supported historically accepted growth curves (Moulton et al., 1922) where the skeleton, indicated by carcass length, was the first to plateau whereas muscle and fat deposition continued to increase in relative proportions. Other correlations among independent variables demonstrated logical allometric growth relationships, as most carcass traits were all positively correlated (P < 0.001) with each other (Table 2). As expected, as HCW increased, loin depth increased (r = 0.41) and loins became heavier (r = 0.64) and longer (stretched loin length, r = 0.41, and compressed loin length, r = 0.45). Also, as HCW increased, fat depth increased (r = 0.30). Previously, Edwards et al. (1981) correlated chilled carcass weight with BF thickness over increasing fat depth ranges (2.29 to 3.28 cm, r = 0.28, and 3.30 to 4.29 cm, r = 0.31), which indicated a stronger relationship with fat thickness as carcasses got heavier. To validate our hypothesis, as carcasses became longer, loins became longer (stretched loin length, r = 0.54, and compressed loin length, r = 0.40). Moreover, carcass length was more correlated (r = 0.67) in carcasses with fat thickness ranging from 4.43 to 5.33 cm than in leaner carcasses (r = 0.46) that ranged in fat thickness from 2.29 to 3.38 cm (Edwards et al., 1981). In the present study, boneless loin weight (r = 0.52) was more correlated to carcass length than LM depth (r = 0.25), even though both relationships were statistically significant (P ≤ 0.001; Table 2). This may be due to changes in the geometric shape in the LM as pigs continue to add lean muscle, therefore causing an underestimation of the actual LM area (LMA; Lowe et al., 2010). Although LM depth is an estimator of LMA; it accounts for only one dimension (depth) and not for muscle width or length. On the other hand, boneless loin weight likely served as a proxy for true dimensional muscle size. Carcass length was historically correlated to backfat depth (r = 0.67; Edwards et al., 1981). However, carcass length was not correlated (r = −0.02, P = 0.58) with fat depth in the present study carcasses (Table 2). The same lack of correlation (r = 0.06) was reported when evaluating the relationship between carcass length and carcass value (Pearson et al., 1970).

View Full Table | Close Full ViewTable 2.

Pearson correlation coefficients (r) of carcass and loin characteristics

 HCW Loin depth Fat depth Boneless loin weight Stretched loin length Compressed loin length Loin depth 0.41* Fat depth 0.30* −0.13* Boneless loin weight 0.64* 0.47* −0.06 Stretched loin length 0.41* 0.14* −0.10 0.40* Compressed loin length 0.45* 0.23* 0.08 0.57* 0.47* Carcass length 0.71* 0.25* −0.02 0.52* 0.54* 0.40*
*P ≤ 0.001.

Still, other carcass traits were related to each other. Loin depth was negatively correlated with fat depth (r = −0.13). This is similar to results from Holland and Hazel (1958), who reported an inverse correlation (r = −0.30) between LMA and backfat thickness. In this population of carcasses, larger loins (greater LM depth) was also correlated with heavier boneless loin weight (r = 0.47) and longer loins (stretched loin length, r = 0.14, and compressed loin length, r = 0.23). Boneless loin weight was correlated with stretched loin length (r = 0.40) and compressed loin length (r = 0.57). Although not as strong as expected, compressed loin length was correlated with (r = 0.47) stretched loin length.

### Coefficients of Determination for Boneless Pork Loin Chop Number

As stated in the introduction, the increase in boneless loin chop price relative to the increase in pork carcass price offers potential for increased revenue from longer pork carcasses compared with shorter carcasses if, in fact, a greater number of boneless loin chops can be cut from loins of longer carcasses. More importantly, measuring carcass length is a noninvasive measurement that can be done quickly at a processing facility and may provide a sorting tool for packers. Unfortunately, carcass length explained only a small portion (R2 = 0.15) of the variation for the number of boneless loin chops produced from a Canadian back loin (Fig. 1). This coefficient of determination aligned with those of previous studies that used carcass length to predict carcass value where carcass length accounted for less than 10% of the variation in lean cutting yields (Cross et al., 1975) and less than 13% of variation in percentage of total lean (Edwards et al., 1981). Hot carcass weight explained 24% (R2 = 0.24) of the variation in the number of chops derived from a boneless loin (Fig. 2). Historically, HCW along with BF depth and LM depth have been used to determine the composition of a pork carcass (Johnson et al., 2004). However, in the present study, a high coefficient of determination was classified as an R2 > 0.60; therefore, under these parameters, carcass weight was a poor predictor of the number and yield of boneless chops.

Figure 1.

Prediction of the number of boneless pork loin chops using carcass length as the independent variable.

Figure 2.

Prediction of the number of boneless pork loin chops using HCW as the independent variable.

The length of a boneless loin can vary, due to the absence of bone to stabilize the muscle. However, the amount of external fat remaining on a boneless loin can affect its ability to compress (Braude et al., 1957). Therefore, determining the minimum length and maximum length of each loin was necessary. The average compressed length was 51.5 cm and the average stretched length was 63.5 cm (Table 1). This equates to an average 18% difference between compressed and stretched lengths for each loin. Compressed loin length was a better predictor (R2 = 0.28) of the number of boneless chops that can be cut from a loin than stretched loin length (R2 = 0.08; Fig. 3 and 4, respectively). Based on the mechanism of the push-feed style slicer used, it is logical that compressed loin length was a better predictor than stretched length, as the loins were likely “semicompressed” as they were sliced. Loin depth explained only 3% (R2 = 0.03) of the variation in the number of boneless loin chops from a boneless loin (Fig. 5). This is not surprising, because geometrically, depth may not be a good predictor of length. From anterior to posterior, LM area, loin width, loin depth, and fat depth all increase; however, the depth:width ratio decreases, indicating that the loin becomes flatter towards the posterior end (Lowe et al., 2010). However, boneless loin weight explained 33% (R2 = 0.33) of the variation in the number of boneless loin chops derived from a boneless loin (Fig. 6). Studies conducted in the 1970s and 1980s intended to use carcass length as an indicator of carcass value because longer carcasses were perceived to be leaner (Braude et al., 1957). Moreover, carcass length will increase with age but, if expressed relative to live or carcass weight, this percentage decreases as the animal gets older (Gerrard and Grant, 2003).

Figure 3.

Prediction of the number of boneless pork loin chops using boneless compressed loin length as the independent variable. Boneless loins (NAMP number 414; NAMP, 2007) were compressed to a minimum length without distortion.

Figure 4.

Prediction of the number of boneless pork loin chops using boneless stretched loin length as the independent variable. Boneless loins (NAMP number 414; NAMP, 2007) were stretched to maximum length without distortion.

Figure 5.

Prediction of the number of boneless pork loin chops using boneless loin depth as the independent variable. Loin depth was determined using a Fat-O-Meater (SFK Technology A/S, Herlev, Denmark) at approximately the area of the 10th rib.

Figure 6.

Prediction of the number of boneless pork loin chops using boneless loin weight as the independent variable.

Even though the number of chops is important, the weight of chops (yield) was also evaluated. Carcass length explained an even lesser portion of the variation in chop yield (R2 = 0.01; Fig. 7). This trend in lack of predictive ability of chop yield using other carcass and loin measurements continued, as HCW explained 5% (R2 = 0.05) of the variation in chop yield. Both boneless loin weight and compressed loin length separately explained 6% (R2 = 0.06) of the variation in chop yield. Stretched loin length and loin depth separately explained only 1% (R2 = 0.01) of the variation in chop yield. Not surprisingly, the weight of the ends and pieces explained the greatest amount of variation in chop yield (R2 = 0.60; Fig. 8). From an economic standpoint, carcass length explained a slightly greater portion of the variation in revenue of total chops per loin compared with chop yield (R2 = 0.25; Fig. 9). Additionally, HCW explained a greater portion of the variation in revenue of total chops per loin compared with carcass length (R2 = 0.40; Fig. 10). This confirms that using weight parameters such as HCW instead of carcass length is a better estimator of the number of chops derived from a boneless pork loin and, therefore, potentially greater revenue from the additional chops. Heavier and longer carcasses do have a greater revenue return; however, carcass weight is a much better predictor than length of the carcass.

Figure 7.

Prediction of chop yield using carcass length as the independent variable.

Figure 8.

Prediction of chop yield using ends and pieces weight as the independent variable.

Figure 9.

Prediction of revenue from total chops per loin using carcass length as the independent variable.

Figure 10.

Prediction of revenue from total chops per loin using HCW as the independent variable.

### Stepwise Regression Model

Carcass length has been used to predict cutability and value. Carcass length was used to predict cutability of the lean cuts (bone-in trimmed ham, loin, picnic shoulder, and Boston butt) but ultimately was a poor predictor (R2 = 0.18) of cutability (Cross et al., 1975). Prior to that, Pearson et al. (1970) attempted to formulate equations to predict whole carcass value systems but concluded that adding carcass length into the stepwise model increased the explanatory power by only 4% in combination with HCW, BF thickness, and LMA. Despite this, there may still be utility in using carcass length to determine the number of chops that can be cut from a boneless loin. Approximately 45% of the variation in the number of boneless loin chops derived from a boneless loin can be accounted for with the following equation: 3.53 + (loin weight, kg × 0.372) + (compressed loin length, cm × 0.113) + (BF depth, mm × 0.050) + (carcass length, cm × 0.034). Carcass length entered the model fourth with a partial R2 = 0.005 (Table 3). The additional marginal R2 value of adding carcass length into the model was minimal and increased the model R2 only from 0.441 to 0.446. This is evident by the conceptual predictive criterion (Cp) for adding carcass length Cp = 16.76. As the Cp approaches the number of parameters, it is indicative of a more unbiased model.

View Full Table | Close Full ViewTable 3.

Summary of stepwise selection of independent variables using the regression procedure of SAS1 to predict the number of 2.54-cm-thick chops derived from a boneless loin

 Step Variable entered Variable removed No. of variables included Partial R2 Model R2 C(p) F value Pr > F 1 Boneless loin weight 1 0.336 0.336 253.84 615.78 <0.0001 2 Compressed loin length None 2 0.070 0.406 101.92 142.34 <0.0001 3 Fat depth None 3 0.035 0.441 25.82 76.73 <0.0001 4 Carcass length None 4 0.005 0.446 16.76 10.95 <0.0001 5 Loin depth None 5 0.004 0.450 9.02 9.72 <0.0001 6 HCW None 6 0.001 0.452 7.85 3.17 <0.0001
1SAS Inst. Inc., Cary, NC.

In addition to chop number, a stepwise regression model for breakeven price (Table 4) and chop yield from each loin (Table 5) was established. When determining a stepwise regression equation for breakeven price, 49.5% of the variation in breakeven price was accounted for with the following equation: −14.827 + (HCW, kg × 0.1246) + (percent lean × 0.2505). With a SLENTRY and SLSTAY level = 0.15, carcass length failed to enter the model for predicting the breakeven price of boneless pork chops from a boneless pork loin. This reaffirms the lack of predictability of carcass length on economic value of a boneless pork loin. For chop yield, 5.4% of the variation was accounted for with the following equation: 85.236 + (HCW, kg × 0.0717) + (percent lean × 0.0713) − (carcass length, cm × 0.0796). Carcass length entered the model second with a partial model R2 = 0.0038. The additional marginal utility of adding carcass length into the model was minimal and increased the model R2 only from 0.0465 to 0.0503.

View Full Table | Close Full ViewTable 4.

Summary of stepwise selection of independent variables using the regression procedure of SAS1 to predict the breakeven price of a boneless loin2

 Step Variable entered Variable removed No. of variables included Partial R2 Model R2 C(p) F value Pr > F 1 HCW 1 0.4031 0.4031 222.589 852.36 <0.0001 2 Percent lean None 2 0.0922 0.4954 1.6261 223.21 <0.0001
1SAS Inst. Inc., Cary, NC.
2Breakeven price was calculated as [(boneless loin weight, kg × boneless center cut loin, strap-off price) − (ends and pieces weight, kg × 72 trim, combo)].

View Full Table | Close Full ViewTable 5.

Summary of stepwise selection of independent variables using the regression procedure of SAS1 to predict chop yield of a boneless loin2

 Step Variable entered Variable removed No. of variables included Partial R2 Model R2 C(p) F value Pr > F 1 HCW 1 0.0465 0.0465 8.18 57.73 <0.0001 2 Carcass length None 2 0.0038 0.0503 5.41 4.75 0.0294 3 Percent lean None 3 0.0039 0.0542 2.5 4.93 0.0266
1SAS Inst. Inc., Cary, NC.
2Chop yield was calculated as [(boneless loin weight, kg − ends and pieces, kg)/boneless loin weight, kg] × 100.

### Conclusions

These data indicate that carcass length is a poor predictor for the number of boneless loin chops yielded from a boneless pork loin and for the associated breakeven cost of each boneless loin. Carcass length explained only approximately 15% of the variation in the number of boneless chops produced by a boneless pork loin. Furthermore, the marginal model R2 utility of adding carcass length to the stepwise regression model was minimal and contributed less than an additional 0.05 units of predictive ability (R2) when loin weight, loin length, and carcass BF depth are also known. Additionally, when predicting chop yield, the same minimal marginal utility of adding carcass length in the model was calculated and for breakeven price of each loin, carcass length did not enter the stepwise model. However, for those markets capturing revenue from chop number rather than chop weight, measuring loin weight, compressed loin length, and BF depth can account for approximately 45% of the variation in the number of chops produced from a boneless pork loin. Therefore, although carcass length is a poor predictor of the number of chops from a boneless loin, the use of loin weight, compressed loin length, and BF depth may aid in predicting the number of chops that can be cut from a boneless loin.