TRUNK REQUIREMENTS IN ALTERNATE ROUTING NETWORKS 299 



Non-coincidence of Busy Hours and Busy Seasons 



With respect to the second question involving the non-coincidence of 

 busy hours among toll centers it should be noted first that trunk recjuire- 

 ments estimated in the nationwide alternate routing trunk study were 

 predicated on a common busy season and a common office busy hour for 

 all toll offices and all intertoll groups. This premise resulted in system- 

 wide requirements which were patently incorrect since it is known that 

 different toll centers have different busy hours and that the busy season 

 for toll traffic in New England for example is in the summer while that 

 for Florida is in the winter. While it was possible to identify the busy 

 season and the average busy hour of each toll center the statistical prob- 

 lem of incorporating such information for some 2,500 toll centers in a 

 completely integrated nationwide study appeared insuperable. 



The seriousness of the error introduced by the above premise is not as 

 great as might at first appear since the New England Company should 

 know the requirements for the busy season of its territory and the 

 Southern Bell Company is equally interested in the busy season require- 

 ments for Florida. It is only in the trunks required for handling traffic 

 between these two areas that a distortion of requirements could be 

 readily demonstrated as a result of assuming the two busy seasons to be 

 coincidental when in fact they are months apart. The example cited is 

 an extreme case which serves to point up the problem. While no evalu- 

 ation has been made of this distortion, and none seems statistically 

 practicable, it is evident that the direction of distortion is toward over- 

 estimation of trunk requirements.* 



Thus it may be confidently stated that the general effect of assuming 

 premise regarding the coincidence of busy seasons and busy hours upon 

 the network as a whole was to compute trunk requirements in some 

 groups more liberally than a precise evaluation of all significant factors 

 would indicate as adequate. Proper evaluation of the effect of non- 

 coincidence of busy seasons and busy hours will likely await the findings 

 of field experience. 



Costs — Alternate Routing Versus No Alternate Routing 



In planning extensive and radical changes in the methods of handling 

 toll traffic on a nationwide basis it was necessary to explore the economic 



* In the New York City .studies previously discussed, a similar assumption was 

 made with respect to the coincidence of busy hours and busy seasons of the local 

 offices. Due to the homogeneity of intra-office traffic the degree of distortion in 

 individual trunk group requirements was considered, except for a very few cases, 

 to be insignificant. 



