I don\'t know if I\'m in the right but here\'s my problem :
I follow bus attendancies with 3 experimental data sources (s1, s2, s3) which give me an estimation of the att