THEORIES FOR TOLL TRAFFIC ENGINEERING IN THE U. S. A. 429 



u 



z 



o 



z 

 I- 



UJ 

 HI 



5 



ul 

 _i 

 _i 

 < 

 o 



U- 



o 



z 

 o 



(- 

 cc 

 o 

 a. 

 o 

 tr 

 a. 



0.001 



12 14 



L = LOAD CARRIED IN ERLANGS 



18 



Fig. 5 — Comparison of NC data on a 16-trunk T-engineered toll group with 

 various load versus NC theories. 



dialed T-engineered group of 16 trunks between Newark, N. J., and 

 Akron, Ohio. Curve A shows the empirically determined "NC encoun- 

 tered" relationship described above for ringdown operation; Curve B 

 gives the corresponding theoretical "NC existing" values. Lines C and D 

 give the operator-dialing results, for morning and afternoon busy hours. 

 The observed points are now seen generally to be significantly above 

 Curve B.* 



At the same time as this change in the "NC encountered" was occur- 

 ring, due to the introduction of operator toll dialing, there seems to have 

 l)een little disturbance to the traditional relationship between load 



* The observed point at 11 erlangs which is clearly far out of agreement with 

 the remainder of the data was produced by a combination of high-trend hours 

 and an hour in which an operator apparently made many re-t^rials in rapid suc- 

 cession. 



