It is evident that these rates are only approxi mately accurate. It will be logical and much more convenient, therefore, to express them in even figures. Furthermore, it will be more convenient if all rates which are nearly alike are grouped together and an average rate taken for the group ; in this manner the number of rates is reduced. This has been done in the above example, and column nineteen gives the final readjusted rates. It will be noted that some of the smaller rates have been increased to make them carry a little more of the burden, and all rates have been rounded off so as to make computations simpler, and also so as to make as few groups of rates as possi ble. If it is desired, these groups may be designated by letters and each unit may be given the letter cor responding to the group in which its rate is found. The time cards sent in from each center may then bear the letter identifying the rate to it; in this way the computation of the expense charge able against each job is greatly facilitated. The oper ation of the plan then becomes identical with the oper ation of any machine rate with the supplementary rate, but obviously the results produced are much more intelligent. This illustration, while it does not pos sess the refinements of the complete plan as outlined by Mr. Church is very useful in showing how the method may be modified to suit small shops, or other plants, where a compromise must be made. With all the modifications and approximations suggested it is evident that costs obtained by this method will be much more equitable than those obtained by the averaging methods.
15. Defects of and objection to production fac tors and supplementary rate.—The manifest advan tages of this method are offset, to some extent, by the great amount of preliminary work and study nec essary to instal the method in plants of large size or complex character. If the 'plant is built and laid out with the cost-finding problem in mind, the work will be greatly lessened, and there is no doubt that this feature of factory management will influence, in no small degree, the construction and arrangement of future industrial plants where many and varied products are to be made. The application to com plex existing plants is difficult, as before noted, and can be accomplished only by a liberal use of good judgment. Nevertheless, as has been pointed out, even these difficulties are not insuperable in most cases.
Moreover, in comparing the accuracy of the dis tribution of expense by this method with that of the averaging methods, it should be noted that some of the production factors of the new method are them selves based on averages. Thus, buildings are not heated and lighted equally all the year round, repairs on all kinds of apparatus are not proportional to elapsed time, and expense material may be bought one day that may not be used for several months.
These variable factors of expense must be spread out over more or less arbitrary periods of time, if costs are to be consistent. Again, the expenses that have been incurred during the time a job has been under construction are not always definitely known. If all jobs could be started on the first day of the month and shipped on the last day, it might be possible to allocate with some degree of accuracy the identical expense incurred by each job. Evidently, such con ditions seldom, if ever, exist. Work must be started whenever necessary, and shipped and billed as soon as possible, and both production factors and supple mentary rates must, in general, be based on previous performances. Usually the record of the previous month is used for rate-setting but some expenses may have to be distributed over longer periods to pre vent costs from being erratic. Any claims for great accuracy made by the advocate of any method of distributing expense should be discounted liber ally, unless the conditions are so simple that the inter pretation of all the factors involved can be clearly seen.
An objection often urged against machine rates, in general, is the penalizing effect on small work which is done on a large machine. It often occurs, especially when work is scarce, that work is done on a machine considerably larger than is actually re quired for the process. Under the machine-rate method of distributing expense, the job so done re ceives a much larger share of expense than if done on a tool exactly suited to the operation; thus the job is penalized more heavily, perhaps, than the market price can bear. Furthermore, most records so made are misleading, if not properly interpreted, and if used for future estimates they are likely to make the quotations too high.
While all this is true, the fact remains that the costs obtained by this machine-rate method are closer to the truth than those obtained by averaging methods, which make no distinction in regard to the varying size and cost of the equipment employed. The ma chine rate is in this respect an index of the efficiency of operation and may serve to call the attention of the manager either to a weakness in his manufactur ing equipment or to the pressing need of more work of certain kinds. This may be particularly true in dull times; for in such times the large machines and processes which normally distribute a large part of the expense because of their higher machine rates are usually the first to become idle as business falls off, and the last to resume operations as business re vives.
This particular feature of factory work, however, is the cause of one of the most difficult of all cost finding problems in factories where there is a wide variety in the cost and size of equipment, and where several lines of product are made. This problem will be discussed in some detail in the next chapter.