A comparison of free-stall barns used by modernized Wisconsin dairies

J. Dairy Sci. 84:528–541 American Dairy Science Association, 2001.

A Comparison of Free-Stall Barns Used by ModernizedWisconsin Dairies

J. Bewley,* R. W. Palmer,* and D. B. Jackson-Smith†University of Wisconsin, Madison 53706*Department of Dairy Science†Program on Agricultural Technology Studies

ABSTRACT

A primary objective of the Wisconsin Dairy Modern-ization Survey was to compare features of free-stallbarns available to dairy producers. This study useddata from a large random sample of expanding dairyfarms to determine whether the theoretical benefits ofparticular free-stall configurations bear out under on-farm conditions. Comparisons were made among herdsusing free-stall barns as their primary housing for newversus remodeled facilities, barn design, bedding used,feed-delivery design, manure removal strategies, ani-mal restraint, maternity areas, overcrowding, and cool-ing methods. Producers who made the transition fromtie-stall housing to free-stall housing were satisfiedwith this decision. New free-stall barns provided a moredesirable environment for the herds than remodeledfree-stall barns, although initial investments werehigher. When new free-stall barns were compared,herds with four-row barns had higher production, lowersomatic cell count, and higher stocking rates than herdswith six-row barns. Respondents were more satisfiedwith four- and six-row barns than with two- and three-row barns. Respondents felt sand provided some advan-tages for cow comfort, while satisfaction with beddingcost and manure handling was higher with mattresses.Dairy Herd Improvement data showed no difference inmilk production or somatic cell count for producers whochose sand or mattress-based free stalls. Respondentswere more satisfied with the use of drive-through feed-ing than other feed-delivery designs. Most producerschose to use tractor scrapers to remove manure; how-ever, producers who used automated systems weremore satisfied with manure management. Few differ-ences were observed when comparing self-locking headgates to palpation rails. Overcrowding did not have anyadverse affect on production or user satisfaction with

Received June 5, 2000.Accepted September 30, 2000.Corresponding author: R. W. Palmer; e-mail: rwpalmer@facstaff.

wisc.edu.

528

feed intake or cow comfort. Using supplemental coolingappeared to facilitate higher production.(Key words: free stall, survey, expansion, housing)

Abbreviation key: FSB = free-stall barn, FTE = fulltime equivalent, RHA = rolling herd average.

INTRODUCTION

The Wisconsin dairy industry is experiencing a majorrestructuring. Dairy operators are incorporating mod-ern technologies to help improve efficiency and the qual-ity of life of their families and workers. Many Wisconsindairy producers have expanded or modernized theiroperations in recent years. Limited research has beenconducted on the real benefits achieved by producerswho have modernized facilities. Speicher et al. (1978)studied difficulties and challenges related to expansionof Michigan dairies in the 1970s and determined thatdairy farmers experienced increased difficulties withanimal health, heat detection, manure handling, andlabor management following expansion. Norell et al.(1981) examined changes in milk production followingexpansion and found that production often drops afterexpansion, depending on the change in housing systemand management. Minnesota research concluded thata common denominator for herds that had increasedmilk production in the early 1990s was a move towardlarger herd sizes and more modern facilities (Stahl etal., 1999). Smith et al. (1997) conducted a survey in1994 to determine the influence of housing on productiv-ity that demonstrated advantages for free-stall barns(FSB), covered feeding, and automatic take-off units.

The Wisconsin Dairy Modernization Survey was de-signed to examine both production responses and pro-ducer perceptions related to the modernization of theiroperation. Surveys that are designed to examine man-agement practices and production responses to man-agement changes are valuable in identifying disparitiesbetween experimental findings and field results (How-ard et al., 1992). The survey used in the present studywas designed to 1) determine what changes were made,2) identify changes in herd productivity resulting fromthese changes, 3) measure producer satisfaction with

OUR INDUSTRY TODAY 529

regard to these changes, 4) provide information to pro-ducers considering future expansion, and 5) identifytopics for further research.

This paper focuses specifically on the use of FSB bya subset of 244 Wisconsin dairies that used such barnsas their primary type of housing. Areas examined in-cluded transition from tie-stall barns, type of barn, newversus remodeled facilities, bedding type, feed-deliverydesign, manure removal, animal restraint, overcrowd-ing, maternity area, and cooling methods.

MATERIALS AND METHODS

Data provided by AgSource and Richland CountyWisconsin Dairy Herd Improvement Cooperative wereutilized to identify Wisconsin herds that had expandedbetween 1994 and 1998. Herds were selected if herdsize had increased by at least 50% for smaller herds(60 to 100 cows) or at least 40% for larger herds (>100cows). Additional herds that had increased herd size,but were not DHI members, were identified via letterswritten to all members of Professional Dairy Producersof Wisconsin. Six hundred and ninety-four surveys weremailed with preaddressed stamped envelopes on April5, 1999. Of the 694 farms that received the survey, 604were identified through DHI, 95 through the Profes-sional Dairy Producers of Wisconsin letter, and 50through other referrals (some herds were listed in mul-tiple sources). Follow-up postcards were sent to thosewho had not responded by April 15. County Extensionagents were asked to encourage dairy producers to com-plete the survey and to offer their assistance if nec-essary.

Our survey was designed to gather information aboutthe use of various management practices and facilitytypes, as well as farmers’ satisfaction with the overalldairy and performance of specific facilities in recentlyexpanded dairies. The survey included 280 questionsrelated to herd size, milking system, housing facilities,cropping and feed-delivery designs, labor management,animal acquisitions, animal handling facilities, and sat-isfaction with the expansion experience.

Three hundred thirty-six (48%) surveys were re-turned by July 16. Because some of the respondentsdid not meet survey criteria, returned unusable ques-tionnaires, or declined to participate, 302 (44%) surveyswere analyzed. Herd summary information from Ag-Source was combined with the survey data when avail-able. Milk production data was only included for Hol-stein herds. Herds were categorized based on currentherd size, magnitude of expansion, and type of expan-sion. Herd size categories were established by dividingherds into quintiles. For many questions, producerswere asked to indicate their satisfaction with a particu-

Journal of Dairy Science Vol. 84, No. 2, 2001

lar aspect of their operation by choosing a number ona scale from one (very dissatisfied) to five (very satis-fied). The authors, with the assistance of survey special-ists, edited data to remove inconsistencies and physicalimpossibilities. In this study, respondents who housedat least 75% of their herd in FSB in 1998 (244 respon-dents) were utilized for FSB comparisons, whereas all302 surveys were utilized for the comparison of herdsthat made the transition from tie-stall barns to FSB.Cows per full-time equivalent (FTE) was calculated bydividing the reported number of milking and dry cowsin 1998 by the number of FTE (1 FTE = 50 h of laborper week).

Statistical analyses were performed using SPSS(1999). Comparisons between groups of herds were con-ducted using an independent sample t-test. Compari-sons involving multiple groups were conducted usingthe general linear model univariate procedure. Differ-ences among means were tested using the least signifi-cant differences post hoc test.

RESULTS AND DISCUSSION

Transition to Free-Stall Barns

Dairy cattle in Wisconsin have traditionally beenhoused in tie-stall or stanchion barns. A paradigm shiftis occurring as more producers are moving to FSB asthe primary housing for the milking herd. Among the244 respondents in this survey that used FSB as theirprimary housing in 1998, 156 had used a tie-stall orstanchion barn exclusively in 1994. This dramaticchange is likely because producers recognize the re-duced labor, increased cow comfort, and increased effi-ciency associated with FSB. Valde et al. (1997) con-cluded that differences in disease incidence and fertilityproblems made FSB a more desirable housing systemthan tie-stall barns.

Changes in production and satisfaction levels for pro-ducers who modernized their operation by changinghousing type were measured in the survey. The groupof producers surveyed observed an average increase inrolling herd average (RHA) for milk of 841 kg (1850lb) during this period. An earlier study (Norrell et al.,1981) reported that producers changing from stall barnto FSB showed reduced milk production immediatelyafter expansion, followed by recovery within 2 yr ofthe housing change. More recent research (Stahl et al.,1999) indicated that producers who switched to FSBhad increased production and improved labor efficiency5 yr after expansion.

The average RHA for milk was 938 kg (2063 lb) higher(P < 0.05) for producers that housed cows exclusivelyin FSB than for producers that housed cows in tie-stallbarns. Some of the difference in production between

BEWLEY ET AL.530

these two housing types could be due to managementpractices of the owners. Producers who use only tie-stall barns are often traditional, smaller, family farmsthat utilize fewer productivity-enhancing technologies(Buttel et al., 2000). Age of facilities may also contributeto this difference, because the tie-stall barns or stan-chion barns were typically older than FSB.

Respondents who changed to free-stall housing weregenerally satisfied with its effect on animal perfor-mance. Survey participants who changed housing typewere asked to compare the performance of their opera-tion in 1994 with its current performance on a scalefrom one (much worse) to five (much better). Producersperceived that housing performance had improved forall factors considered. The percentages of producerswho selected “much better” for housing performancemeasures were as follows: feeding convenience, 91%;cow comfort, 81%; hock damage, 74%; teat damage,71%; cow cleanliness, 57%; ability to move animals,57%; feed intake, 53%; udder health, 48%; and manuremanagement, 40%. Group feeding in FSB, rather thanindividual feeding of animals in tie-stall barns, proba-bly had a major influence on the responses regardingease of feeding. Free-stall design, bedding material se-lection, and animal freedom contributed to the highsatisfaction ratings for cow comfort, cow cleanliness,hock damage, and teat damage. Manure managementhad the lowest increase in level of satisfaction and wasprobably related to a change from barn cleaners in tie-stall barns to tractor scraping in FSB since tractorscraping increases manure handling labor re-quirements.

New Versus Remodeled Free-Stall Barns

The majority (66%) of producers in this study utilizednew FSB exclusively, and the remainder used eitherremodeled FSB (10%) or both new and remodeled FSB(23%). Table 1 shows production levels and satisfactionaverages for respondents in these three categories. Av-erage RHA for milk in 1998 did not differ between thethree groups, although the increase in average RHAfor milk since 1994 was higher (P < 0.05) for herds usingremodeled FSB (2936 vs. 1655 and 1745 lb; respec-tively). This suggests that both new and remodeledfacilities can support high production if constructed cor-rectly. The producers who remodeled their facilities didnot increase herd size as dramatically and had lowerinvestment per stall than producers in the other twocategories.

Respondents who used new FSB or new and remod-eled FSB had higher (P < 0.05) satisfaction with cowcomfort, cow cleanliness, and feeding convenience thanrespondents using only remodeled facilities. Respon-


dents who had new FSB were more satisfied with theirability to move animals and manage manure than re-spondents with remodeled facilities. The differences incow comfort and cow cleanliness and ability to moveanimals likely occurred because producers who builtnew FSB were not hindered by the constraints of anexisting facility. Seventy-five percent of producers withnew FSB utilized drive-through feeding, and this maycontribute to the difference in satisfaction with feedingconvenience. All of the satisfaction parameters favoredthe construction of new housing; however, this must beexamined considering the costs associated with eachchoice. Reported cost per stall for respondents usingremodeled FSB was less (P < 0.05) than those who chosenew and remodeled or new only ($534 vs. $980 and$1,107; respectively).

Free-Stall Type

Respondents to this survey elected to build barnsin the following proportions: four-row, 39%; three-row,25%; six-row, 23%; and two-row, 7%. The disadvantagesfor three- and six-row barns compared with two- andfour-row barns are increased heat stress and reducedfeeding area (Smith et al., 1996), although buildingspace requirements should be lower for these barns.The three-row and six-row barns also require three rowsof cows on two manure alleys versus two rows of cows;which increases cow congestion and may restrict accessto feed and water. A six-row barn is wider and resultsin about 20% less building space per cow than a four-row barn on a per-stall basis. As a result, ventilationcan become a problem because more cows share a givenamount of airflow (Midwest Plan Service, 1997).

Free-stall barn configurations were compared for allFSB barns (all FSB) and for respondents with only newFSB to remove the effect of remodeled structures (Table2). When all FSB were compared, the average RHA formilk in 1998 was higher (P < 0.05) for four-row barnsthan for three-row barns (22,938 vs. 21,528 lb). Amongnew FSB, 1998 RHA for milk was higher (P < 0.05) forfour-row barns than for two-, three-, and six-row barns(23,455 vs. 20,482, 21,297, and 22,077 lb; respectively).Increases in production during the 5-yr period studiedamong the four groups were not significantly different.Average linear SCS was significantly lower (P < 0.05)in new four-row barns than in new six-row barns (2.71vs. 2.95).

Operations with six-row barns achieved more cowsper FTE than respondents with two- or three-row barns(47 vs. 35 and 38; respectively). Among only new FSB,cows per FTE was higher for four- and six-row barnsthan for two- and three-row barns. Producers who builttwo- and three-row barns had smaller herds. This differ-


Table 1. Mean (± SE) production and satisfaction measures by age of free-stall barns.

Remodeled1 Both2 New3

Herds, no. 24 55 1601998 median herd size 127 245 2211994 median herd size 75 100 82Stocking rate,4 % 107 108 1081998 RHA5 milk, lbs 22,507 ± 615a 22.735 ± 468a 22,258 ± 258a

1994 RHA5 milk, lbs 19,570 ± 600b 21,079 ± 457a 20,513 ± 252ab

Change in RHA,6 lbs 2936 ± 440a 1655 ± 335b 1745 ± 185b

Linear SCS 2.81 ± 0.10a 2.88 ± 0.08a 2.82 ± 0.05a

Cows per FTE7 40 ± 4a 39 ± 2a 44 ± 1a

Cow comfort8 4.13 ± 0.16b 4.53 ± 0.11a 4.54 ± 0.06a

Cow cleanliness8 4.00 ± 0.16b 4.44 ± 0.11a 4.40 ± 0.06a

Hock damage8 4.75 ± 0.16a 4.53 ± 0.11a 4.52 ± 0.06a

Teat damage8 4.75 ± 0.16a 4.60 ± 0.10a 4.50 ± 0.06a

Udder health8 4.42 ± 0.13a 4.24 ± 0.11a 4.25 ± 0.06a

Feed intake8 4.21 ± 0.17a 4.32 ± 0.11a 4.37 ± 0.06a

Ability to move animals8 3.89 ± 0.19b 4.22 ± 0.12ab 4.34 ± 0.07a

Bedding usage and cost8 4.29 ± 0.19a 4.13 ± 0.13a 3.99 ± 0.07a

Manure management8 3.29 ± 0.22b 3.75 ± 0.15ab 3.84 ± 0.09a

Feeding convenience8 3.79 ± 0.19b 4.40 ± 0.13a 4.66 ± 0.07a

Cost per stall, $ 534 ± 150b 980 ± 68a 1107 ± 38a

a,bMeans within rows with different superscripts differ (P < 0.05).1Remodeled = Converted shed or remodeled existing free-stall barn.2Both = Converted shed or remodeled existing free-stall barn and built new free-stall barn.3New = Built new free-stall barn.4Stocking rate = Number of cows/number of stalls.5RHA = DHI calculated Rolling Herd Average.6Change in RHA = 1998 RHA Milk − 1994 RHA Milk.7Cows per FTE = Number of milking and dry cows/(total hours worked per wk/50).8Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).

ence in labor efficiency may be partially related to theircrop enterprise. The average acres per cow were 3.38,3.37, 2.64, 2.61, and 2.31 for herds of the 60 to 105, 106to 145, 146 to 220, 221 to 360, and >360 cows per herd,respectively. The three largest groups had fewer (P <0.05) acres per cow than the two smaller groups.

No differences in producer satisfaction were observedamong the four FSB types for hock damage, teat dam-age, udder health, feed intake, bedding usage and cost,or manure management. Producer satisfaction withtwo-row barns was lower than for other types of barnsfor cow comfort, cow cleanliness, and feeding conve-nience when all FSB were evaluated, but not for newFSB. Remodeled facilities may have contributed to thisdifference, and this difference may be due to outsidefeeding or some other design feature. Satisfaction withthe ability to move animals was lower for three-rowbarns than for other FSB, and this may be due to diffi-culty in moving cows from back pens through front penswith this design. Reported cost per stall was higher forfour-row and six-row barns than for two- and three-rowbarns, because narrower buildings are cheaper to buildand drive-by feeding reduces barn size requirements.When only new drive-through FSB were compared (Ta-ble 3), the cost per stall was not significantly different.The cost of six-row barns may have been influenced by


the fact that 20% of these barns were built with slattedfloors compared with only 2% of four-row barns.

Both four- and six-row barns are normally built withdrive-through feeding, and most large-scale producerscontemplating a new FSB consider four- or six-row de-signs. Data from Table 5 indicate the overall superiorityof a drive-through feed-delivery design versus otheralternatives. To remove the effects of feed delivery de-sign and remodeled barns, we compared free-stall con-figurations among the subset of respondents who hadnew FSB and utilized drive-through feeding (Table 3).The results show that 1998 RHA for milk was higher(P < 0.05) in four-row barns than three-row or six-rowdesigns (23,644 vs. 21,379 and 21,733 lb; respectively),while having a higher (P < 0.05) stocking rate (112 vs.95 and 103%; respectively). No significant difference inRHA was observed prior to expansion. Average linearSCS was lower (P < 0.05) in four-row barns than in six-row barns (2.73 vs. 2.96). No significant differences,among this subset of farms, in the levels of producersatisfaction, labor efficiency, or cost per stall werefound.

Bedding Type

With FSB, the stall base and bedding should be de-signed to supply a comfortable bed with a clean, dry

BEWLEY ET AL.532

Tab

le2.

Mea

n(±

SE

)pr

odu

ctio

nan

dsa

tisf

acti

onm

easu

res

ofdi

ffer

ent

type

sof

free

-sta

llba

rns.

All

free

-sta

llba

rns1

New

free

-sta

llba

rns2

2-ro

w3-

row

4-ro

w6-

row

2-ro

w3-

row

4-ro

w6-

row

Her

ds,

no.

1760

9655

840

5946

1998

med

ian

her

dsi

ze14

413

024

032

013

312

624

529

819

94m

edia

nh

erd

size

8169

9411

578

7087

114

Sto

ckin

gra

te,3

%10

410

311

110

410

110

411

210

219

98R

HA

4m

ilk,

lbs

22,2

91±

806ab

21,5

28±

407b

22,9

38±

329a

22,4

69±

484ab

20,4

82±

1064

b21

,297

±47

6b23

,355

±41

1a22

,077

±51

4b

1994

RH

A4

mil

k,lb

s19

,861

±77

9ab19

,765

±39

3b20

,994

±31

8a20

,852

±46

8ab18

,729

±10

43b

19,7

07±

466b

21,4

85±

402a

20,5

99±

503ab

Ch

ange

inR

HA

5 ,lb

s24

31±

585a

1764

±29

5a19

44±

239a

1617

±35

2a17

53±

789a

1591

±35

3a19

69±

305a

1478

±38

1a

Lin

ear

SC

S2.

82±

0.14

a2.

87±

0.07

a2.

78±

0.06

a2.

97±

0.08

a3.

09±

0.19

ab2.

83±

0.08

ab2.

71±

0.08

b2.

95±

0.10

a

Cow

spe

rF

TE

635

±4b

38±

2b44

±2ab

47±

2a31

±6b

38±

3b47

±2a

47±

3a

Cow

com

fort

74.

00±

0.19

b4.

67±

0.10

a4.

54±

0.08

a4.

50±

0.10

a4.

13±

0.29

a4.

69±

0.13

a4.

56±

0.11

a4.

48±

0.12

a

Cow

clea

nli

nes

s73.

88±

0.19

b4.

50±

0.10

a4.

33±

0.08

a4.

39±

0.10

a4.

13±

0.27

a4.

46±

0.12

a4.

32±

0.10

a4.

46±

0.11

a

Udd

erh

ealt

h7

4.12

±0.

19a

4.35

±0.

10a

4.31

±0.

08a

4.17

±0.

10a

4.38

±0.

26a

4.36

±0.

12a

4.19

±0.

10a

4.20

±0.

11a

Fee

din

take

74.

29±

0.19

a4.

21±

0.10

a4.

40±

0.08

d4.

35±

0.10

a4.

38±

0.28

a4.

26±

0.13

a4.

39±

0.10

a4.

37±

0.12

a

Abi

lity

tom

ove

anim

als7

4.12

±0.

21ab

4.11

±0.

12b

4.30

±0.

09ab

4.57

±0.

12a

4.50

±0.

30ab

4.13

±0.

14b

4.29

±0.

11ab

4.59

±0.

13a

Bed

din

gu

sage

and

cost

74.

12±

0.22

a3.

95±

0.12

a4.

18±

0.09

a4.

09±

0.12

a4.

25±

0.35

a3.

80±

0.16

a4.

10±

0.13

a4.

00±

0.14

a

Man

ure

man

agem

ent7

3.82

±0.

26a

3.67

±0.

14a

3.81

±0.

11a

3.94

±0.

15a

4.00

±0.

36a

3.72

±0.

16a

3.88

±0.

13a

3.87

±0.

15a

Fee

din

gco

nve

nie

nce

73.

82±

0.22

b4.

53±

0.12

a4.

55±

0.09

a4.

74±

0.12

a4.

50±

0.28

a4.

59±

0.13

a4.

68±

0.10

a4.

76±

0.12

a

Cos

tpe

rst

all,

$71

2±

120b

908

±64

b11

20±

53a

1215

±65

a80

4±

155b

904

±74

b11

99±

64a

1206

±68

a

a,b M

ean

sw

ith

inro

ws

wit

hdi

ffer

ent

supe

rscr

ipts

diff

er(P

<0.

05).

1 All

free

-sta

llba

rns

=P

rodu

cers

repo

rtin

gh

avin

gn

ew,

rem

odel

ed,

orbo

thty

pes

offr

ee-s

tall

barn

s.2 N

ewfr

ee-s

tall

barn

s=

On

lypr

odu

cers

repo

rtin

gh

avin

gn

ewfr

ee-s

tall

barn

s.3 S

tock

ing

rate

=N

um

ber

ofco

ws/

nu

mbe

rof

stal

ls.

4 RH

A=

DH

Ica

lcu

late

dro

llin

gh

erd

aver

age.

5 Ch

ange

inR

HA

=19

98R

HA

mil

k−

1994

RH

Am

ilk.

6 Cow

spe

rF

TE

=N

um

ber

ofm

ilki

ng

and

dry

cow

s/(t

otal

hou

rsw

orke

dpe

rw

eek/

50).

7 Ave

rage

sati

sfac

tion

repo

rted

ona

scal

efr

om1

(ver

ydi

ssat

isfi

ed)

to5

(ver

ysa

tisfi

ed).



Table 3. Mean (± SE) production and satisfaction measures of different types of new, drive-through free-stall barns.

2-row 3-row 4-row 6-row

Herds, no. 3 16 53 421998 median herd size 163 116 245 2751994 median herd size 81 63 87 114Stocking rate,1 % 99ab 95b 112a 103b

1998 RHA2 milk, lb 21,503 ± 1758ab 21,379 ± 718b 23,644 ± 379a 21,733 ± 478b

1994 RHA2 milk, lb 19,267 ± 1927ab 19,798 ± 787b 21,669 ± 416a 20,351 ± 524ab

Change in RHA3, lb 2237 ± 1499a 1581 ± 612a 1974 ± 323a 1382 ± 408a

Linear SCS 3.00 ± 0.32ab 2.83 ± 0.13ab 2.73 ± 0.07b 2.96 ± 0.09a

Cows per FTE4 24 ± 11b 37 ± 5ab 47 ± 3a 48 ± 3a

Cow comfort5 3.67 ± 0.47b 4.47 ± 0.21ab 4.66 ± 0.11a 4.45 ± 0.12ab

Cow cleanliness5 3.67 ± 0.45a 4.27 ± 0.20a 4.36 ± 0.11a 4.43 ± 0.12a

Udder health5 4.33 ± 0.45a 4.07 ± 0.20a 4.19 ± 0.11a 4.19 ± 0.12a

Feed intake5 3.67 ± 0.47a 4.07 ± 0.21a 4.47 ± 0.11a 4.33 ± 0.13a

Ability to move animals5 4.33 ± 0.50ab 3.73 ± 0.22b 4.34 ± 0.12a 4.60 ± 0.13a

Bedding usage and cost5 3.67 ± 0.56ab 3.53 ± 0.25b 4.13 ± 0.13a 4.02 ± 0.15ab

Manure management5 4.00 ± 0.60a 3.67 ± 0.27a 3.91 ± 0.16a 3.95 ± 0.16a

Feeding convenience5 4.67 ± 0.44a 4.53 ± 0.20a 4.76 ± 0.11a 4.76 ± 0.12a

Cost per stall, $ 1094 ± 307a 1003 ± 120a 1235 ± 71a 1212 ± 76a

a,bMeans within rows with different superscripts differ (P < 0.05).1Stocking rate = Number of cows/number of stalls.2RHA = DHI calculated rolling herd average.3Change in RHA = 1998 RHA milk − 1994 RHA milk.4Cows per FTE = Number of milking and dry cows/(total hours worked per week/50).5Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).

Table 4. Mean (± SE) production and satisfaction measures of different free-stall bedding types.

All free-stall barns1 New free-stall barns2

Mattresses Sand Mattresses Sand

Herds, no. 69 145 51 911998 median herd size 265 195 257 2001994 median herd size 110 80 87 80Stocking rate3, % 106 108 106 1071998 RHA4 milk, lb 22,519 ± 420a 22,539 ± 264a 22,508 ± 495a 22,293 ± 341a

1994 RHA4 milk, lb 20,931 ± 411a 20,468 ± 258a 21,050 ± 480a 20,378 ± 330a

Change in RHA5, lb 1587 ± 0.07a 2071 ± 189a 1458 ± 345a 1915 ± 237a

Linear SCS 2.88 ± 0.07a 2.80 ± 0.05a 2.84 ± 0.09a 2.79 ± 0.06a

Cows per FTE6 45 ± 2a 40 ± 1a 47 ± 3a 41 ± 2a

Culling rate (%) 34 ± 2a 32 ± 1a 34 ± 1a 29 ± 1b

Cow comfort7 4.42 ± 0.10a 4.55 ± 0.07a 4.35 ± 0.11b 4.65 ± 0.08a

Cow cleanliness7 4.12 ± 0.10b 4.47 ± 0.07a 4.10 ± 0.10b 4.59 ± 0.08a

Hock damage7 4.22 ± 0.09b 4.72 ± 0.06a 4.18 ± 0.11b 4.76 ± 0.08a

Teat damage7 4.48 ± 0.09a 4.59 ± 0.07a 4.45 ± 0.11a 4.59 ± 0.09a

Udder health7 4.09 ± 0.10a 4.31 ± 0.07a 4.08 ± 0.10b 4.35 ± 0.08a

Bedding usage and cost7 4.25 ± 0.11a 3.95 ± 0.08b 4.22 ± 0.13a 3.86 ± 0.10b

Manure management7 4.32 ± 0.12a 3.43 ± 0.08b 4.28 ± 0.13a 3.53 ± 0.10b

Cost per stall, $ 1306 ± 58a 946 ± 41b 1312 ± 67a 1021 ± 49b

a,bMeans within rows with different superscripts differ (P < 0.05).1All free-stall barns = Producers reporting having new, remodeled, or both types of free-stall barns.2New free-stall barns = Only producers reporting having new free-stall barns.3Stocking rate = Number of cows/number of stalls.4RHA = DHI calculated rolling herd average.5Change in RHA = 1998 RHA milk − 1994 RHA milk.6Cows per FTE = Number of milking and dry cows/(total hours worked per week/50).7Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).


BEWLEY ET AL.534

Table 5. Mean (± SE) production and satisfaction measures of different types of feed-delivery designs.

All free-stall barns1 New free-stall barns2

Drive-through1 Drive-by4 Outside feeding5 Drive-through3 Drive-by4

Herds, no. 155 44 19 119 271998 median herd size 245 144 125 245 1351994 median herd size 90 70 85 85 70Stocking rate,6 % 107 107 103 106 1081998 RHA7 milk, lb 22,657 ± 269a 21,608 ± 460ab 20,981 ± 741b 22,620 ± 292a 21,254 ± 559b

1994 RHA7 milk, lb 20,941 ± 259a 19,711 ± 443b 18,719 ± 714b 20,865 ± 294a 19,622 ± 563ab

Change in RHA8, lb 1715 ± 196a 1897 ± 335a 2262 ± 540a 1755 ± 218a 1632 ± 417a

Cows per FTE9 45 ± 12a 37 ± 3b 40 ± 4ab 46 ± 2a 37 ± 4a

Linear SCS 2.84 ± 0.05a 2.90 ± 0.08a 2.93 ± 0.13a 2.82 ± 0.05b 2.90 ± 0.10b

Cow comfort10 4.55 ± 0.06a 4.55 ± 0.12a 4.11 ± 0.18b 4.54 ± 0.07ab 4.82 ± 0.15a

Cow cleanliness10 4.38 ± 0.06a 4.39 ± 0.12a 4.05 ± 0.18a 4.38 ± 0.07a 4.52 ± 0.15a

Feed intake10 4.38 ± 0.07a 4.14 ± 0.12a 4.26 ± 0.19a 4.34 ± 0.07a 4.41 ± 0.15a

Ability to move animals10 4.38 ± 0.07a 4.02 ± 0.14b 4.16 ± 0.21ab 4.36 ± 0.08a 4.37 ± 0.17a

Manure management10 3.94 ± 0.09a 3.55 ± 0.16b 3.26 ± 0.24b 3.91 ± 0.09a 3.74 ± 0.20a

Feeding convenience10 4.73 ± 0.07a 4.36 ± 0.14b 3.74 ± 0.21c 4.74 ± 0.07a 4.67 ± 0.15a

Cost perstall, $ 1196 ± 37a 877 ± 69b 522 ± 117c 1197 ± 44a 896 ± 92b

a,b,cMeans within rows with different superscripts differ (P < 0.05).1All free-stall barns = Producers reporting having new, remodeled, or both types of free-stall barns.2New free- stall barns = Only producers reporting having new free-stall barns.3Drive-through = Free-stall barn with feed delivery apron inside the barn.4Drive-by = Free-stall barn with feed delivery apron on the side of the barn.5Outside feeding = Cows walk outside of free-stall barn to eat.6Stocking rate = Number of cows/number of stalls.7RHA = DHI calculated rolling herd average.8Change in RHA = 1998 RHA milk − 1994 RHA milk.9Cows per FTE = Number of milking and dry cows/(total hours worked per week/50).10Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).

surface (Midwest Plan Service, 1997; Stowell, 2000).The most common choices for stall bases among surveyrespondents were sand (59%) and mattresses (28%). Asand bed provides both the base and the bedding for astall. Sand has been promoted as a stall base becauseit is inorganic, drains well, provides traction in alleys,is less likely to harbor mastitis-causing organisms, andis less conducive to hock injuries (Stowell, 2000). Sandhas lower bacterial levels, which can influence mastitis(Britten, 1994; Hogan et al., 1989; McFarland et al.,1994; Stowell, 2000). Bedding mattresses are oftenplaced over a hard stall base such as concrete. Mat-tresses provide adequate cushion for the cow, and bed-ding costs are often reduced. Research indicates thatcows prefer mattresses to other materials placed overa concrete base, as indicated by frequency and durationof use (Rodenburg et al., 1994, 2000). Supplementalbedding in the form of sawdust, straw, corn stalks,shredded newspaper, or other organic material is re-quired to keep the mattress surface dry.

We found that the RHA for milk and linear SCS didnot differ (P < 0.05) between herds using mattressesand herds using sand (Table 4). Smith et al. (1997) alsoreported that free-stall bedding (sand, dirt, sawdust,shavings, peanut hulls, or mattresses) did not signifi-cantly affect milk quality. Although no differences in


linear SCS were observed among the herds in thisstudy, SCS may not fully reflect incidences of environ-mental mastitis because of the short duration of theseinfections. Producers with sand reported higher satis-faction (P < 0.05) with udder health than those withmattresses. This may reflect differences in mastitis ex-perienced with inorganic bedding materials and produc-ers’ perceptions of the incidence of environmental mas-titis in the herd.

In other research, cow comfort has been measuredas a function of stall usage. That research (Rodenburget al., 2000) suggested that mattresses provided greatercow comfort than did sand. Producers have reported(Britten, 1994) reduced cow comfort and increased ma-nure-handling problems when inorganic bedding mate-rials are used. Rodenburg et al. (2000) also indicatedthat hock abrasions were less prevalent in sand-beddedFSB compared with mattresses.

The following cow comfort comparisons can be madebased on producer perceptions. In new FSB, herds usingsand had lower (P < 0.05) culling rates than herds usingmattresses (34 vs. 29%). Producers using sand weremore satisfied (P < 0.05) with cow cleanliness than pro-ducers using mattresses. Producers using sand in newFSB reported higher (P < 0.05) satisfaction with cowcomfort. Respondents using sand were more satisfied


(P < 0.05) with the amount of hock damage than respon-dents using mattresses. Sand provides more tractionfor animals and shifts easily as the cow rises from thestall, so cows are less likely to slip and injure theirhocks.

Producers using mattresses were more satisfied (P <0.05) with bedding usage and cost and manure manage-ment than producers who used sand. Sand can accumu-late rapidly in the manure reception and storage pitsand is abrasive to equipment and difficult to pump. Asa result, producers often have trouble handling sand-laden manure. The present study indicates that bothsystems can support comparable milk production andSCS.

The cost of FSB with mattresses was higher (P < 0.05)than the cost of FSB with sand bases. The purchaseand installation of mattresses added to the expenseof constructing a FSB. Respondents who chose to usemattresses were more likely to have new facilities, four-row or six-row barns, drive-through feeding, and auto-mated manure collection systems. Each of these factorscontributes to the difference in cost discussed above.This survey indicates a trend to a lower labor require-ment and decreased bedding usage and cost with mat-tress-based barns, but no attempt was made to calculatethe ongoing costs of sand and mattress-based barns.Additional research is needed to determine the annu-alized costs of each.

Responses related to changing bedding types werealso observed. The two most predominant changes wereswitching from straw to sand and from straw to mat-tresses. Survey participants who changed beddingtypes were asked to compare the performance of their1994 operation to that of their current operation ona scale from one (much worse) to five (much better).Producers who made the change from straw bedding tosand reported the change in average performance forudder health was better than those who made thechange from straw to mattresses. This could indicatea reduction in environmental mastitis for herds switch-ing to inorganic bedding material. The group thatswitched stall base but still used organic bedding mate-rials did not perceive as great an improvement in ud-der health.

The percentages of producers who switched fromstraw to sand and who selected a “much better” perfor-mance, were as follows: cow comfort, 90%; hock damage,88%; teat damage, 82%; cow cleanliness, 71%; feed in-take, 66%; udder health, 63%; ability to move animals,52%, bedding usage and cost, 43%, and manure man-agement, 32%. Improvements in cow comfort, teat dam-age, cow cleanliness, and udder health are likely relatedto the inorganic properties of sand and the additionaltraction and cushion provided. The percentages of pro-


ducers who switched from straw to mattresses and whoselected “much better” for housing performance mea-sures were as follows: cow comfort, 80%; teat damage,74%; manure management, 71%; ability to move ani-mals, 69%; hock damage, 63%; cow cleanliness, 51%;bedding usage and cost, 49%. Improvements in cowcomfort, teat damage, and hock damage are related todesign features of mattresses. Improved manure man-agement may be related the smaller amount of beddingmaterial needed on top of mattress-based free stalls.

Feed-Delivery Design

Three different feeding-delivery designs were exam-ined in this survey: drive-through feeding, where feedis delivered inside the barn (64%); drive-by feeding,where feed is delivered along a side of the barn (18%);and outside feeding, where animals must go outside toeat (8%). Among all FSB in this study, producers whoused drive-through feeding had higher (P < 0.05) aver-age RHA for milk in 1998 than respondents who usedoutside feeding (Table 5). No differences were observedamong feed-delivery designs for change in RHA for milkover the period studied. Among only new FSB, produc-ers with drive-through feeding had higher (P < 0.05)average RHA for milk in 1998 than respondents withdrive-by feeding, although the change in RHA for milkwas not significant. A study of southern US dairy herds(Smith et al., 1997) showed that herds fed inside theFSB or under a covered roof had higher milk productionand lower SCS than those fed outside. Production indrive-by barns may be lower than in drive-throughbarns because cow comfort may be compromised andfeed may be damaged by precipitation.

For all FSB, respondents using drive-through feedinghad higher cows per FTE than respondents using drive-by feeding. Respondents using drive-through feedingreported higher (P < 0.05) satisfaction with feeding con-venience. Drive-through feeding is more desirable forthe operator, especially when feed is pushed up, becauseof the efficient arrangement of the barn and the coverprovided by indoor feeding. Respondents using drive-through feeding had higher (P < 0.05) satisfaction withmanure management. Seventy-eight percent of produc-ers using drive-through feeding had new facilities, anddifferences in manure handling observed were likelyassociated with age of facilities rather than feed-deliv-ery design. Another possible explanation for this advan-tage in manure handling is that four-row barns andsix-row barns are designed to handle manure from fourpens with one collection point. Respondents using drive-through feeding reported higher (P < 0.05) satisfactionwith their ability to move animals. This may be theresult of better cow traffic patterns when planning a

BEWLEY ET AL.536

new FSB. Cow traffic patterns can be considered inthe planning process for new FSB more easily than inremodeled FSB. No differences in cow traffic satisfac-tion were observed when only new FSB were con-sidered.

Free-stall barns with drive-through feeding cost moreper stall than FSB with drive-by feeding, and FSB withdrive-by feeding cost more per stall than FSB with out-side feeding. These differences in costs were a functionof differences in age of the barns and the proportion offour-row barns and six-row barns using drive-throughfeeding. The cost of a four-row barn or a six-row barnis typically higher than two-row or three-row barnsbecause of the additional width of the barn and theadditional square footage required for covering the feed-ing lane. Some producers using outside feeding mayhave calculated the cost of the barn without includingthe cost of the outside feeding area.

Manure Removal

Manure management becomes increasingly im-portant as herds increase in size. Frequent, effectivemanure removal is necessary to keep cows clean. Cleancows are less susceptible to mastitis and require lesspreparation time in the parlor. Producers often mustchoose between labor-intensive tractor scraping or morecapital-intensive automated manure collection systemssuch as automatic scrapers, slatted floors, and flushsystems. The total costs and time associated with auto-mated manure collection are frequently lower than fordaily tractor scraping (Graves, 1994). The frequency ofmanure removal methods in this survey were (Table6): tractor scrape (80%), alley scrapers (11%), slattedfloors (7%), and flush systems (2%), indicating a largepercentage of these producers chose the less capital-intensive approach.

Respondents using alley scrapers, in all FSB and newFSB, achieved higher cows per FTE than respondentsusing tractor scraping (Table 6). With cows per FTE as ameasure of labor efficiency, producers who used slattedfloors and flush systems had higher labor efficiencythan producers who used tractor scraping, but insuffi-cient numbers were available to show a statistical dif-ference.

Examining both all FSB and new FSB, respondentsusing alley scrapers, slatted floors, or flush systemswere more satisfied with manure management thanrespondents using a tractor to scrape manure. Tractorscraping was probably selected by producers to decreaseinitial costs, but the increased labor and managementrequirements of tractor scraping decreased their la-bor efficiency.


Tab

le6.

Mea

n(±

SE

)pr

odu

ctio

nan

dsa

tisf

acti

onm

easu

res

ofm

anu

re-r

emov

alsy

stem

s.

All

free

-sta

llba

rns1

New

free

-sta

llba

rns2

Tra

ctor

scra

peA

lley

scra

pers

Sla

tsF

lush

Tra

ctor

scra

peA

lley

scra

pers

Sla

tsF

lush

Her

ds,

no.

189

2617

511

720

134

1998

med

ian

her

dsi

ze20

528

337

054

520

538

024

059

019

44m

edia

nh

erd

size

8110

511

530

080

113

9528

3S

tock

ing

rate

,3%

107

108

106

109

106

112

105

108

Cow

spe

rF

TE

442

±1b

50±

3a43

±5ab

38±

10ab

43±

2b55

±4a

44±

6ab44

±13

ab

Man

ure

man

agem

ent5

3.55

±0.

07b

4.39

±0.

20a

4.65

±0.

25a

5.00

±0.

45a

3.62

±0.

09b

4.25

±0.

21a

4.54

±0.

27a

5.00

±0.

48a

Bed

din

gu

sage

and

cost

53.

95±

0.07

b4.

39±

0.18

a4.

41±

0.22

a4.

20±

0.41

ab3.

85±

0.09

b4.

35±

0.21

a4.

31±

0.27

ab4.

00±

0.48

ab

Cos

tpe

rst

all,

$98

6±

37b

1111

±99

b14

58±

112a

1095

±24

2ab10

43±

45b

1161

±10

8ab14

15±

121a

1106

±29

6ab

a,b M

ean

sw

ith

inro

ws

wit

hdi

ffer

ent

supe

rscr

ipts

diff

er(P

<0.

05).

1 All

free

-sta

llba

rns

=P

rodu

cers

repo

rtin

gh

avin

gn

ew,

rem

odel

ed,

orbo

thty

pes

offr

ee-s

tall

barn

s.2 N

ewfr

ee-s

tall

barn

s=

On

lypr

odu

cers

repo

rtin

gh

avin

gn

ewfr

ee-s

tall

barn

s.3 S

tock

ing

rate

=N

um

ber

ofco

ws/

nu

mbe

rof

stal

ls.

4 Cow

spe

rF

TE

=N

um

ber

ofm

ilki

ng

and

dry

cow

s/(t

otal

hou

rsw

orke

dpe

rw

eek/

50).

5 Ave

rage

sati

sfac

tion

repo

rted

ona

scal

efr

om1

(ver

ydi

ssat

isfi

ed)

to5

(ver

ysa

tisfi

ed).


Among all FSB, respondents using alley scrapers orslatted floors were more satisfied with bedding usageand cost than respondents using a tractor to scrapemanure. This advantage was only observed for alleyscrapers versus tractor scraping in new FSB. A highpercentage (72%) of respondents with tractor scrapingused sand for bedding, and this relationship betweenselection of tractor scraping and sand bedding may ex-plain the difference in satisfaction with bedding usageand cost for manure-collection methods. Producers withmore capital-intensive manure-removal methods likelychose mattresses because they are more compatiblewith automated manure-collection systems.

For all FSB, the slatted-floor system was significantlymore costly per stall than alley scrapers or tractorscraping. This difference was only observed for slatted-floor barns over tractor scraping in new FSB. Buildinga pit under a barn adds to the initial construction costs.Miller (1998) compared projected costs of slatted-floorbarns to barns designed for tractor scraping and foundthat, although investment costs for the slatted-floor sys-tem were 13.6% higher, 10 yr costs per cow were 68.3%lower. No health-related costs were included in theseprojections.

The majority (79%) of producers who changed fromone type of manure handling system to another duringthe 5 yr period studied used barn cleaners in 1994 andtractor scraping in 1998. These producers also changedtheir housing type from tie-stall barns to FSB. Surveyparticipants who changed manure handling strategieswere asked to compare the performance of their 1994operation to that of their current operation using ascale from one (much worse) to five (much better). Thepercentage of respondents reporting “much better” forchange in average performance were as follows: hockdamage, 78%; cow cleanliness, 62%; bedding usage andcost, 41%; and manure management, 36%.

Animal Restraint

Proper handling of animals requires the ability tosort and restrain animals for pregnancy diagnosis,treatments, estrus synchronization, AI, and hoof trim-ming. Self-locking head gates include a mechanism thatallows for the catch and release of all cows along afeeding line. Palpation rails have been introduced re-cently as an inexpensive alternative to self-locking headgates when handling a large number of cows. Hardinet al. (1994) claimed that advantages associated withpalpation rails include simplicity, cost, and construc-tion ease.

Respondents who used self-locking head gates or pal-pation rails, in combination with other methods (animalhandling chutes, accessing cows when in milking par-


Table 7. Mean (± SE) satisfaction values for animal restraint systemsin free-stall barns with either palpation rails or self-locking headgates.

Self-lockinghead gates Palpation rails

Herds, no. 102 271998 median herd size 349 2761994 median herd size 129 981998 RHA1 milk, lb 22,962 ± 283a 23,244 ± 841a

1994 RHA1 milk, lb 21,205 ± 308a 21,290 ± 803a

Change in RHA milk,2 lb 1787 ± 249a 1954 ± 540a

Cows per FTE3 45 ± 2a 44 ± 4a

Feed intake4 4.38 ± 0.07a 4.31 ± 0.20a

Initial cost4 3.71 ± 0.11b 4.73 ± 0.12a

Labor efficiency4 4.51 ± 0.07a 4.38 ± 0.20a

Ease of use/comfort of worker4 4.52 ± 0.07a 4.42 ± 0.19a

Worker safety4 4.40 ± 0.08a 4.54 ± 0.17a

a,b,cMeans within rows with different superscripts differ (P < 0.05).1RHA = DHI calculated rolling herd average.2Change in RHA = 1998 RHA milk − 1994 RHA milk.3Cows per FTE = Number of milking and dry cows/(total hours

worked per week/50).4Average satisfaction reported on a scale from 1 (very dissatisfied)

to 5 (very satisfied).

lor, catching animals in parlor return lane, and corner-ing animals in pen or free-stall), were compared (Table7). Respondents using palpation rails were more (P <0.05) satisfied with the initial cost of animal restraintsystems than respondents using self-locking headgates. The investment in building palpation rails in-volves a simple post-and-rail system in a centralizedlocation. Self-locking head gates require investmentsfor multiple self-locking mechanisms, ideally one forevery cow in the barn.

Batchelder (2000) reported that feed intake is sig-nificantly lower for cows in barns with self-locking headgates than it is for cows in barns without self-lockinghead gates, although animals in this study were onlyprovided 1 wk to adjust to the self-locks. Bolinger et al.(1997) demonstrated that milk yield, SCC, DMI, andmastitis were not affected by restraint with self-lockswhen cows were locked up for less than 4 consecutivehours. Although some people are concerned that self-locking head gates reduce intake, this study shows nodifferences between the two groups for milk productionor user satisfaction with DMI.

Based on these data, the type of animal restraintdoes not seem to affect the overall labor efficiency ofthe operation, because no differences were observed forcows per FTE or user satisfaction with labor efficiency.Satisfactions with ease of use, worker safety, andworker comfort for self-locking head gates or palpationrails were not statistically different. Some (Hardin etal., 1994) have claimed that operator safety is a benefitof palpation rails.

BEWLEY ET AL.538

Maternity

To properly manage a modern dairy herd, a maternityarea should be provided to house and handle cows dur-ing calving to reduce stress and manage disease trans-mission. A separate, disinfected maternity area is a keyto a successful fresh cow and replacement managementprogram (Heinrichs et al., 1987). The primary calvingalternatives studied in this survey were individual calv-ing pens and bedding packs. A Pennsylvania study (Hei-nrichs et al., 1987) showed that fewer than 25% of herdshad any type of maternity pen, and many of these wereconsidered inadequate. In our survey, almost 76% ofherds with free-stall housing had most or all of theiranimals calve in individual pens or on a bedding pack.Individual calving pens are more expensive to buildthan a bedding pack and require more labor.

Responses from producers who indicated that mostor all of their animals calved in one particular locationwere compared (Table 8). Respondents using individualcalving pens reported higher (P < 0.05) satisfaction withcalf health than those using a bedding pack. This con-curs with previous research showing that calf mortalityis lower in herds when cows calve in individual boxstalls compared with those calving in group maternitypens (Speicher et al., 1986). Calves born on beddingpacks are likely exposed to higher concentrations ofdisease-causing organisms than calves born in individ-ual calving pens. Disinfecting between calvings mayalso be more prevalent with individual calving pens.Calving pens are generally viewed as providing ahealthier environment; however, in this research nodifferences were observed for satisfaction with cow

Table 8. Mean (± SE) satisfaction values for calving locations in free-stall barns using respondents who indicated that “most” or “all” oftheir cows calved in the respective location.

Individual Group calving oncalving pens bedding pack

Herds, no 80 1051998 median herd size 200 2451994 median herd size 81 1001998 RHA1 milk, lb 22,784 ± 344a 22,701 ± 310a

1994 RHA1 milk, lb 20,777 ± 345a 20,763 ± 308a

Change in RHA milk,2 lb 2039 ± 258a 1938 ± 230a

Cows per FTE3 40 ± 2b 46 ± 2a

Cow health4 3.93 ± 0.11a 3.69 ± 0.10a

Calf health4 4.04 ± 0.11a 3.60 ± 0.10b

Labor use4 3.83 ± 0.12a 3.63 ± 0.11a

a,bMeans within rows and columns with differ superscripts differ(P < 0.05).

1RHA = DHI calculated rolling herd average.2Change in RHA = 1998 RHA milk − 1994 RHA milk.3Cows per FTE = Number of milking and dry cows/(total hours

worker per week/50).4Average satisfaction reported on a scale from 1 (very dissatisfied)

to 5 (very satisfied).


health. Respondents using a bedded pack achievedhigher (P < 0.05) cows per FTE than those using individ-ual calving pens, but producer satisfaction with laboruse did not differ between the two groups. The increasein cows per FTE with bedded peaks probably is a resultof easier monitoring and reduced handling of cows.

Overcrowding

Overcrowding a barn is a typical solution to increas-ing cow numbers without increasing facility invest-ments. Concerns over cow comfort, reduced DMI, andlowered milk production discourage many producersfrom this practice. Virginia researchers (Friend et al.,1977) suggested that a barn can be overcrowded byup to 30% without affecting production or behavior.Cornell researchers (Batchelder, 2000) demonstratedthat fewer animals eat postmilking and a lower percent-age chew their cud in an overcrowded barn than in abarn without overcrowding. These lower chewing timeswere associated with increased acidosis and increasedlameness. A British study (Leonard et al., 1996) indi-cated that overcrowded animals may have increasedincidence of lameness resulting from reduced lyingtimes.

Stocking rates for herds in this study with new FSBwere calculated by dividing the reported number ofcows housed in FSB by the reported number of stalls.Herds were divided into overcrowding categories to ex-amine the effects of different stocking rates. No differ-ences were observed between categories for satisfactionwith DMI (Table 9). In this survey, average RHA formilk in 1998 and change in average RHA for milk dur-ing the 5-yr period studied were not statistically differ-ent between stocking rates. Satisfaction with cow com-fort was consistent across all overcrowding categories.

No differences were observed for FSB cost per stall,but cost per cow decreased with increasing stockingrates and was lowest for barns that were 21 to 30%overcrowded. For all FSB, average stocking rates of111% for four-row and 104% for six-row barns werereported (Table 2). The same production trends werefound when four-row barns and six-row barns were ana-lyzed separately. Few observations were available forsix-row barns that were overcrowded, making statisti-cal comparisons more difficult.

Cooling Methods

The majority (57%) of respondents did not use fansor sprinklers (Table 10), 24% used fans only, 7% usedsprinklers only, and 11% used both fans and sprinklers.Cows should be provided with good air movement (Mid-west Plan Service, 1997). Proper ventilation designed


Table 9. Mean (± SE) production and satisfaction measures of different barn stocking rates in new free-stall barns.

1–10% 11–20% 21–30% >30%Under capacity At capacity overcrowded overcrowded overcrowded overcrowded

Herds, no 46 4 46 37 16 81998 median herd size 128 115 263 275 373 2301994 median herd size 70 54 89 90 110 801998 RHA1 milk, lb 21,501 ± 469a 20,693 ± 1464a 22,743 ± 502a 22,768 ± 544a 22,244 ± 976a 23,297 ± 1035a

1994 RHA1 milk, lb 19.621 ± 446b 19,123 ± 1393ab 21,075 ± 478a 21,128 ± 517a 20,032 ± 928ab 21,682 ± 985ab

Change in RHA,2 lb 1879 ± 334a 1570 ± 1042a 1668 ± 358a 1640 ± 387a 2211 ± 695a 1615 ± 737a

Cow comfort3 4.49 ± 0.11a 5.00 ± 0.38a 4.57 ± 0.11a 4.65 ± 0.13a 4.44 ± 0.19a 4.50 ± 0.27a

Feed intake3 4.33 ± 0.11a 5.00 ± 0.38a 4.35 ± 0.11a 4.43 ± 0.12a 4.50 ± 0.19a 4.13 ± 0.27a

Cost per stall, $ 1076 ± 70a 1183 ± 217a 1162 ± 73a 1174 ± 91a 916 ± 120a 1192 ± 194a

Cost per cow, $ 1266 ± 82a 1183 ± 254abc 1094 ± 86ab 988 ± 106bc 708 ± 141c 793 ± 228abc

a,b,cMeans within rows with different superscripts differ (P < 0.05).1RHA = DHI calculated rolling herd average.2Change in RHA = 1998 RHA milk − 1994 RHA milk.3Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).

to improve air movement is especially important in thesummer. Supplemental cooling fans are used to forceair over cows for more rapid heat loss. Sprinkling sys-tems are often used to reduce the effects of extremetemperatures on cows. Excessive temperatures do notallow an animal to produce to her genetic potential. Astudy of Florida cows (Strickland et al., 1988) found thatcows housed with fans and sprinklers had significantlyhigher DMI and milk production than those withoutfans and sprinklers during the summer.

Average RHA for milk in 1998 was higher (P < 0.05)for herds using fans, sprinklers, or fans and sprinklers,although no differences were observed when looking atchange in average RHA for milk over the 5-yr period.Any long-term effects of heat stress, such as decreasedreproductive efficiency, decreased immunity, or in-creased mastitis, were not studied.

No differences were observed for average linear SCS.Cow comfort satisfaction scores did not vary statisti-cally between groups. Respondents using neither fansnor sprinklers were more satisfied (P < 0.05) with cow

Table 10. Mean (± SE) production and satisfaction measures for producers using fans and sprinklers.

Fans andsprinklers Fans Sprinklers Neither

Herds, no. 28 49 18 1391998 median herd size 243 283 203 1901994 median herd size 125 95 83 801998 RHA1 milk, lb 22,964 ± 419a 23,943 ± 857a 23,381 ± 636a 21,800 ± 266b

1994 RHA1 milk, lb 20,841 ± 418a 21,309 ± 855a 20,993 ± 634a 20,183 ± 267a

Change in RHA milk,2 lb 2123 ± 305a 2634 ± 624a 2388 ± 463a 1627 ± 195a

Linear SCS 2.82 ± 0.07a 2.71 ± 0.14a 2.78 ± 0.12a 2.87 ± 0.05a

Cow comfort3 4.42 ± 0.11a 4.33 ± 0.19a 4.29 ± 0.15a 4.56 ± 0.07a

Cow cleanliness3 4.14 ± 0.10b 4.39 ± 0.19ab 4.32 ± 0.15ab 4.43 ± 0.07a

a,bMeans within rows with different superscripts differ (P < 0.05).1RHA = DHI calculated rolling herd average.2Change in RHA = 1998 RHA milk − 1994 RHA milk.3Average satisfaction reported on a scale from 1 (very dissatisfied) to 5 (very satisfied).


cleanliness than respondents using both fans and sprin-klers. Producers with both fans and sprinklers reportedless satisfaction with cow cleanliness than those withneither, which can probably be attributed to wet stallsor cows lying in manure alleys.

CONCLUSIONS

Producers who made the transition from tie-stall toFSB appear to be satisfied with this decision. Highproduction can be supported by either new or remodeledfacilities, although producers appear to be more satis-fied with new facilities. Overall, four-row barns andsix-row barns outperform two-row barns and three-rowbarns. Drive-through feeding is superior to drive-by oroutside feeding for almost all factors measured. Withnew drive-though FSB, producers choosing four-rowbarns had higher stocking rates, higher milk produc-tion, and lower average SCC than producers who chosesix-row barns. Data from DHI showed no difference inmilk production or SCC for producers who chose sand

BEWLEY ET AL.540

or mattress-based free-stalls. Producers who chose sandhad higher cow comfort satisfaction scores, but mat-tresses were preferred for bedding usage, cost, and ma-nure management issues. Alley scrapers, slatted floors,and flush systems appear to allow for easier manuremanagement than tractor scraping. No differences wereobserved between self-locking head gates and palpationrails for labor efficiency, although producers were moresatisfied with the initial cost of palpation rails. Individ-ual calving pens appear to provide an advantage forcalf health. Overcrowding did not affect milk productionfor herds in this survey. Supplemental cooling improvesmilk production but appears to reduce cow cleanliness.

The results of this survey can be used by industryprofessionals and producers who are considering mod-ernizing their operations. Our use of large random sam-ple survey data identified some impacts associated withchanges in facilities or management practices that dif-fer from those expected (based on the results of pre-viously published case studies, controlled experimentsand engineering models). These differences reflect theeffects of variation in farmer management ability, envi-ronmental and economic conditions, and other intangi-ble factors. The experiences of farmers in our surveycan be combined with other types of research data toprovide a more complete picture of how modernizationchoices affect dairy farm performance. Further researchis merited regarding production differences betweenfree-stall types, levels of overcrowding, and differencesbetween animal handling systems.

ACKNOWLEDGMENTS

The authors would like to thank the producers whoparticipated in the survey for the valuable informationprovided, AgSource DHI and Richland County DHICfor herd summary data, the Professional Dairy Produc-ers of Wisconsin for producer names, and the Programon Agricultural Technology Studies for assistance indesigning the survey. The authors also thank DougHemken, Social Science Microcomputer Laboratory, forhis assistance in the statistical analysis of the data.This research has been sponsored in part by USDA/Hatch project number 1892.

REFERENCES

Batchelder, T. L. 2000. The impact of head gates and overcrowdingon production and behavior patterns of lactating dairy cows.Pages 325–330 in Dairy Housing and Equipment Systems: Man-aging and Planning for Profitability, Camp Hill, PA. NaturalResource, Agric., and Engineering Service, Ithaca, NY.

Bickert, W. G. 1998. Sorting, handling, and restraining lactating cowsfor treatment and other purposes. Pages 44–50 in Proc. 4th Int.Dairy Housing Conf., St. Louis, MO. Am. Soc. Agric. Eng., St.Joseph, MI.


Bolinger, D. J., J. L. Albright, J. Morrow-Tesch, S. J. Kenyon, andM. D. Cunningham. 1997. The effects of restraint using self-locking stanchions on dairy cows in relation to behavior, feedintake, physiological parameters, health, and milk yield. J. DairySci. 80:2411–2417.

Britten, A. M. 1994. Dairy free stall bedding systems and udderhealth. Pages 292–299 in Proc. Natl. Mastitis Counc., Natl. Masti-tis Counc., Inc, Arlington, VA.

Buttel, F. H., D. Jackson-Smith, and S. Moon. 2000. A Profile ofWisconsin’s Dairy Industry, 1999. PATS Res. Rep. No. 3. Programon Agricultural Technology Studies, Univ. Wisconsin, Madison.

Friend, T. H., C. E. Polan, and M. L. McGilliard. 1977. Free stalland feed bunk requirements relative to behavior, production andindividual feed intake in dairy cows. J. Dairy Sci. 60:108–116.

Grant, R. 1997. Cow comfort and herd health: a nutrionist’s perspec-tive. Pages 351–360 in Proc. Western Canadian Dairy Seminar,Red Deer, AB. Univ. of Alberta, Edmonton, AB, Canada.

Graves, R. E. 1994. Dairy manure handling. Pages 263–266 in Expan-sion Strategies for Dairy Farms, Ellicott City, MD. Penn. St.Univ., Harrisburg, PA.

Graves, R. E. 1994. Restraint and treatment facilities for dairy ani-mals. Pages 34–38 in Expansion Strategies for Dairy Farms,Ellicott City, MD. Penn. St. Univ., Harrisburg, PA.

Hardin, D. K., R. F. Randle, and J. M. Zulovich. 1994. Creating apalpation workstation to enhance reproductive herd health. Pages637–639 in Proc. 3rd Int. Dairy Housing Conf., Orlando, FL. Am.Soc. Agric. Eng., St. Joseph, MI.

Heinrichs, A. J., R. E. Graves, and N. E. Kiernan. 1987. Survey ofcalf and heifer housing on Pennsylvania dairy farms. J. DairySci. 70:1952–1957.

Hogan, J. S., K. L. Smith, K. H. Hoblet, D. A. Todhunter, P. S.Schoenberger, W. D. Hueston, D. E. Pritchard, G. L. Bowman,L. E. Heider, B. L. Brockett, and H. R. Conrad. 1989. Bacterialcounts in bedding materials used on nine commercial dairies. J.Dairy Sci. 72:250–258.

Howard, W. H., R. W. Blake, T. O. Knight, C. R. Shumway, and M.A. Tomaszewski. 1992. Estimating the effectiveness of extensioninformation systems using farm trials and subjective probabili-ties. Agric. Econ. 7:77–90.

Leonard, F. C., J. M. O’Connell, and K. J. O’Farrell. 1996. Effect ofovercrowding on claw health in first-calved Friesian heifers. Br.Vet. J. 152:459–472.

McFarland, D. F., and M. J. Gamroth. 1994. Free-stall design withcow comfort in mind. Pages 270–285 in Proc. 33rd Annu. Mtg.Natl. Mastitis Counc., Natl. Mastitis Counc., Inc., Arlington, VA.

Midwest Plan Service. 1997. Dairy Free-stall Housing and Equip-ment. 6th ed. Iowa State Univ. Press, Ames, IA.

Miller, P. 1998. Cost comparison of slatted floor and scrape manuresystems. Pages 363–369 in Proc. 4th Int. Dairy Housing Conf.,St. Louis, MO. Am. Soc. Agric. Eng., St. Joseph, MI.

Norrell, R. J., and R. D. Appleman. 1981. Change of milk productionwith housing system and herd expansion. J. Dairy Sci.64:1749–1755.

Rodenburg, J., and H. K. House. 2000. The impact of free-stall baseand design on cow comfort. Pages 214–224 in Dairy Housing andEquipment Systems: Managing and Planning for Profitability,Camp Hill, PA. Natural Resource, Agric., and Engineering Ser-vice, Ithaca, NY.

Rodenburg, J., H. K. House, and N. G. Anderson. 1994. Free stallbase and bedding materials: effect on cow comfort. Pages 286–291 in Proc. 33rd Annu. Mtg. Natl. Mastitis Counc., Natl. MastitisCounc., Inc., Arlington, VA.

Rushen, J., and A. M. de Passille. 1999. Environmental design forhealthier and more profitable cows. Pages 319–333 in Proc. ofthe Western Canadian Dairy Seminar, Red Deer, AB. Univ. ofAlberta, Edmonton, AB, Canada.

Smith, J. F., and J. Harner. 1998. Planning a Dairy Expansion. MF-2318. Kansas State Univ. Agric. Experiment Station and Coopera-tive Extension Service, Manhattan, KS.

Smith, J. W., and L. O. Ely. 1997. The influence of feeding and housingsystems on production, reproduction, and somatic cell countscores of Southern Holstein herds. Prof. Anim. Sci. 13:155–161.


Smith, K. L., and J. S. Hogan. 2000. Bedding’s contribution to mastitisin dairy cows. Pages 265–270 in Dairy Housing and EquipmentSystems: Managing and Planning for Profitability, Camp Hill, PA.Natural Resource, Agric., and Engineering Service, Ithaca, NY.

Speicher, J. A., S. B. Nott, and T. L. Stoll. 1978. Changes in produc-tion, cash flow, and income with dairy herd expansion. J. DairySci. 61:1242–1249.

Speicher, J. A., and W. G. Bickert. 1986. Relationship of free stallhousing to management. Pages 85–91 in Dairy Free Stall HousingSymp., Harrisburg, PA. Northeast Regional Agric. EngineeringService, Ithaca, NY.

SPSS Base 10.0 User’s Guide. 1999. SPSS Inc. Chicago, IL.Stahl, T. J., B. J. Conlin, A. J. Seykora, and G. R. Steuernagel.

1999. Characteristics of Minnesota dairy farms that significantly


increased milk production from 1989 to 1993. J. Dairy Sci.82:45–51.

Stowell, R. R. 2000. Sand for bedding. Pages 226–234 in Dairy Hous-ing and Equipment Systems: Managing and Planning for Profit-ability, Camp Hill, PA. Natural Resource, Agric., and EngineeringService, Ithaca, NY.

Strickland, J. T., R. A. Bucklin, D. K. Beede, and D. R. Bray. 1988.Sprinkler and fan evaporative cooling for dairy cattle in Florida.Am. Soc. Agric. Eng. Paper No. 88-4042. Am. Soc. Agric. Eng.,St. Joseph, MI.

Valde, J. P., D. W. Hird, M. C. Thurmond, and O. Osteras. 1997.Comparison of ketosis, clinical mastitis, somatic cell count, andreproductive performance between free stall and tie-stall barns inNorweigan dairy herds with automatic feeding. Acta Vet. Scand.38:181–192.

Documents

A comparison of free-stall barns used by modernized Wisconsin dairies