562 



BELL SYSTEM TECHNICAL JOURNAL 



percentage of NC existing as determined from the Erlang "B" formula 

 (Fig. 14). Using Table T-1 as an example, the NC conditions are as follows; 



Similar comparisons made with the other capacity tables indicate similar 

 uniformity in the most frequently used portion of the tables, i.e., up to 20 or 

 30 trunks. The results are as follows: 



Conclusion 



Since the primary function of an intertoll trunk capacity table is to 

 translate a desired speed of toll service into the number of trunks required 

 for that level of service, the table used should be indicative, within reason- 

 able limits, of the probable effect of trunk provision on the overall speed. 

 For this reason, tables which reflect a uniform service situation will be more 

 useful in intertoll trunk engineering and administration than the present 

 tables which have inherent service variations. Capacity tables such as 

 Tables T-1 to T-5 will therefore be substituted for present Schedules A, A2 

 and B. 



The author gratefully acknowledges the helpful cooperation of those in the 

 several Associated Companies who participated in collection of the empirical 

 data. Thanks are also extended to A. S. Mayo for his guiding hand; to 

 R. I. Wilkinson and F. F. Shipley for their helpful comments; to K. W. 

 Halbert and Miss C. A. Lennon who computed the necessary extensions of 

 the PoUaczek formula; and to Miss E. B. Schaller for her skill in preparing 

 the numerous curves. 



