Hur påverkas kotrafiken av blandfodrets sammansättning vid utfodring i automatiska mjölkningssystem?

Studiens syfte var att undersöka hur sammansättningen av blandfoder påverkar den frivilliga kotrafiken (hur ofta korna frivilligt går till mjölkning) på gårdar med automatisk mjölkning (AM). Studien genomfördes på elva gårdar i Västra Götalands, Gävleborgs och Västernorrlands län där korna utfodrades med blandfoder på foderbordet totalt 939 kor. Den enskilda kons frivilliga mjölkningsfrekvens, mjölkavkastning samt laktationsstadium m.m. registrerades under två perioder på alla mjölkande kor, tillsammans med blandfodrets sammansättning och egenskaper samt antal kor per robot. Under den första perioden gick korna på sin ordinarie foderstat och under den andra förändrades sammansättningen av mixen på foderbordet. Den totala näringsförsörjningen förändrades dock ej mellan perioderna för att ej sammanblanda effekten av ändrad utfodringsnivå med effekten av förändringen av mixens sammansättning. Ändringen i mixen mellan första och andra perioden syftade till att uppnå en höjning eller sänkning av robotkvoten, (definierat som stärkelse+restkolhydrater / NDF (fiber)). Det rekommenderas att robotkvoten ska vara under 0,6 för en fungerande kotrafik. Foderstaterna beräknades i Typfoder (NorFor) och informationen om alla individuella kor togs från datorn kopplad till AM-enheten på gårdarna. Resultatet visade att fyllnadsvärdet (hur mycket fodret fyller våmmen) hade en positiv effekt på kotrafiken. Fyllnadsvärdets ökning från 0,39 till 0,46 gav i studien en signifikant ökning av +0,2 frivilliga mjölkningar per dag. Robotkvoten hade i denna studie endast en signifikant effekt på den frivilliga mjölkningsfrekvensen hos kor i tidig laktation (<90 dagar i mjölk). Resultatet visade också att en högre energinivå i fullfodermixen hade en negativ effekt på antalet frivilliga mjölkningar per ko och dag. Antalet kor per robot hade en starkt signifikant effekt på antalet frivilliga mjölkningar per dag, ca - 0,1 mjölkningar per dag när antalet kor ökade med ca 5 kor (beräknat inom intervallet 41-74 kor). På samma sätt minskade mjölkningsfrekvensen över laktationen och en ko i tidig laktation hade i genomsnitt en högre mjölkningsfrekvens jämfört med en ko senare i laktationen, en minskning på ca -0,2 frivilliga mjölkningar per 100 dagar längre fram i laktationen.The aim of the study was to investigate how composition of a feed mix affects voluntary milking frequency when offered to cows in a system with automatic milking (AM) with special emphasis on the effect of the so-called “robotic quotient” on milking frequency. The robotic quotient is defined as (starch+rest carbohydrates)/neutral detergent fiber. The value is used when giving advice to farmers who have problems with low milking frequencies in the robot and a robotic quotient below 0.6 is often recommended to maintain high milking frequencies. In the set used, voluntary milking frequency was defined as milking frequency minus the number of times a cow was fetched to the milking robot. The study was conducted on 939 cows on eleven farms with AM-systems offered partial mixed ration (PMR) or total mixed ration (TMR) on the feed table. Registrations of each cow’s individual voluntary milking frequency was recorded during two periods along with information about stage of lactation, lactation number, breed etc. Composition of each feed mix offered on the feed table was determined by feed sampling, weighing of constituents and feed analyses, components were calculated according to the Nordic feed evaluation system (NorFor) and data were registered along with farm data about the number of cows per robot, robot brand, number of concentrate feeders etc. In the first period the standard feeding on each farm was recorded and thereafter, at the start of period 2 a change in the feeding regime on each farm was introduced, with the aim of changing the robotic quotient plus or minus 0.1 unit. The change introduced depended on the standard feeding regime and on the feeds that were available on each farm leading to an increase on some farms and a decrease on others. The changes in the composition of the mix offered on the feed table introduced at the start of the second period were accompanied by modifications of the amount of concentrates in the robot, in order to maintain the same supply of nutrients to the cows over both periods, so that the effects of the PMR/TMR composition would not be confounded with the effect of nutrient supply. Data was analyzed statistically where the effect of feed constituents on voluntary milking frequency was evaluated in models that also included animal and herd factors if statistically significant The results showed that changing the robot quotient had an effect (P<0.05) on cow traffic for cows in early lactation (0-90 days in milk). The factor that proved to give the best model for explaining voluntary milking frequency was the fill value, a variable calculated in the NorFor system to describe how much the feed physically fills the animal. The fill value proved to have a positive effect on cow traffic and an increase in fill value from 0.39 to 0.46 increased the number of voluntary milkings with +0.2 per day (P<0.001). An increase in net energy content of the ration had a negative effect on the number of voluntary milkings per cow and day (P<0.01). In a similar manner metabolisable energy content of the silage had a negative effect on voluntary milking frequency (P<0.05). Among animal and herd factors, the number of cows per robot proved to have a strong effect (P<0.001) on milking frequency. Within the studied interval of 41-74 cows per robot, a decrease of 0.1 milkings per day with an increase of 5 cows per robot gave a decrease of 0.1 milkings per day. Stage of lactation affected voluntary milking frequency with a decrease of 0.2 milkings per day with +100 days in milk (P