550 



BELL SYSTEM TECHNICAL JOURNAL 



this factor by observed data should the conditions governing the use of 

 alternate routes change substantially or should the actual results at some 

 future time on groups provided with alternates be found to differ from those 

 currently predicted. 



Delays are expressed in Hgs. 5 and 6 as a percentage of the holding time 

 per average use of the trunk. The holding time factors used in intertoll 

 trunk engineering generally are expressed in terms of the holding time 

 per message. Therefore, in order to use the curve conveniently it is neces- 

 sary to reduce the holding time per message to the holding time per trunk 



UJ 

 HCt 4 



ZQ- 



Suj 4 



_|Z 



UJQ 



C^-i 3, 



ujZ 2 



.9 1 



< 



2.0 



8 9 10 n 

 SIZE OF GROUP 



Fig. 7 — Average circuit delay on all calls resulting from present intertoll trunk capacity 

 schedules (without alternate routes)-^Using Pollaczek formula. 



use or attempt. This can be done with sufficient accuracy for this purpose 

 by a ratio of 1.5 outward attempts per call disposed of. 



Description of Engineering Tables 



Having determined the average circuit delay, as previously described, the 

 delays which result from the present Intertoll Trunk Capacity Tables A, 

 A2, and B can be determined by referring the percentages of use on these 

 tables to the cur\'es in Figs. 5 and 6. Figure 7 indicates that each of these 

 existing tables results in variable delays depending upon the size of the 

 group. 



