THIS PAPER DERIVED FROM AN INVESTIGATION, THE AIM OF WHICH WAS TO DEVELOP A SYSTEM OF COUNTING ON SOUTH AUSTRALIAN ROADS, AND OF PROCESSING THIS DATA, WHICH WOULD PERMIT AN INCREASE IN THE NUMBER OF LOCATIONS AT WHICH AADT ESTIMATES COULD BE TAKEN, AS WELL AS INCREASING THE ACCURACY OF THESE ESTIMATES WITHOUT INCREASING THE MAN HOURS OF WORK REQUIRED. BASED ON A METHOD DERIVED BY HARMELINK (1968) THIS PAPER FITS VARIOUS MODELS TO FULL YEAR COUNT DATA IN AN ATTEMPT TO FOLLOW THIS DATA WITH A 'SMOOTH' CURVE. SEVERAL 'SHORT TERM' SAMPLE COUNTING OPTIONS ARE THEN TESTED TO DETERMINE WHAT ACCURACY THEY WILL GIVE IN ESTIMATING AADT. THE WORK WAS CARRIED OUT BY SIMULATING TRAFFIC SITUATIONS FROM EXISTING FULL YEAR COUNT DATA AND COMPARING ESTIMATED AADTS WITH THE KNOWN TRUE VALUE. (A). FOR THE COVERING ABSTRACT OF THE CONFERENCE PROCEEDINGS SEE IRRD ABSTRACT NO. 226354.
Samenvatting