It is evident that this modern method is by far the most ,economical, and that fewer tools of any given kind will be needed, since the possibility of keeping all machines in operation is much greater when all tools of one kind, and the work which can be done upon them, are collected in one place. The indirect expense for superintendence will be less, and the volume of the product will be greater, since special ization can be made more effective.
Both principles, organization on the basis of prod ucts, and that on the basis of processes, should be given careful attention in arranging the plant, espe cially if cost finding is being considered. In a very large works, for instance, the entire plant may be ar ranged by products, certain definite buildings being devoted to definite lines of work, so that the total costs of each line may be easily segregated. The tools and processes used in each line of work may, however, be arranged to best advantage by the other method; that is, according to the processes performed. Careful consideration of the arrangement may make it pos sible to use a simple cost-finding system, whereas lack of such consideration may make the problem so com plex that it cannot be solved even by the most elabo rate method. It should, of course, be noted that strict adherence to either form of organization is neither necessary nor indeed always possible. Many depart ments are equipped primarily to perform a given proc ess, often requiring a few tools of entirely different kinds, simply to save time and transportation. While space does not permit further discussion of this prob lem, it is one which requires the careful attention of the manager who is interested in costs.
6. Résumé of methods of distributing expense.— The consideration of the several methods of distri buting expense and the discussion in Chapters X and XI regarding the character of its several items, will have made it clear that exact allocation of factory burden is not, in general, possible. It is true that in very simple cases, such as are found in the continuous industries, the distribution may be as accurate as is desired, but the general case of intermittent manu facturing does not admit of an exact solution. Even tho clerical machinery be installed which, theoretically, will give accurate results, the expense items them selves are so variable as to make exact distribution difficult even in simple cases, and impossible where the conditions are complex. This fact is made clear when
it is considered that the exact expense to be distributed is itself not always accurately known; distribution must be based on records of the past week, month, or year, as the case may be. In most cases the record of the previous month is used as a basis, but it may be that in some cases it will be better to base the distribu tion on the average of a number of preceding months. In all cases, however, heavy periodic expense invest ments should be spread out so as to equalize the costs. The fact that there are so many methods and theories of distributing expense is perhaps due, to some extent, to the fact that there is no absolutely accurate method. The need of some theory on which to build a system of expense distribution is self-evident, but it is equally plain that the theory adopted should be capable of practical application. The method that will be satis factory in one case may not do at all in another. There are, however, two fundamental principles that should be remembered no matter what system of dis tribution is being considered.
First, the cost records obtained should state, so far as possible, the facts of the case; they should be rec ords of events that have happened and of nothing else. A clear-cut distinction should be made between costs, as such, and their interpretation and use, which is an other matter. Systems of cost based on estimates are in common use. Obviously, such methods cannot be considered seriously, either for accuracy or as a guide for future action. But, altho the primary object of a cost system to state facts, due consideration should be given to the use to which these facts are to be put. In general, the two main uses of costs are, first, to show how money has been spent and, second, to indi cate how such expenditures may be controlled. In organizing a cost-finding system, therefore, the form in which cost returns are recorded should be carefully considered so that classification and analysis may be rendered easy and simple. A later chapter will treat this matter more fully.