watch to vary in the average time consumed in performing certain process. This information is good so far as it goes, but for best results the time study should be detailed and

Figure 28. Detailed Time Study of Two Operators Labeling Packages By this method of presentation the reader may see clearly the relative length of time for different operations as well as the comparison of total time taken by the two workers. Dimension marks and figures show conveniently, the actual number of seconds required. The different operations have been given numbers instead of names. The scale to which the chart is drawn is named. (Reproduced by permission from Brinton's Graphic Methods for Presenting Facts.)

the findings presented graphically. This has been done in Figure 28. Since it is often desired to present graphically in its component parts the disposition of the company's gross revenues, the relative amount sold of various supplies in a service station, or the items entering into the cost of a commodity, the horizontal or vertical bar has here a considerable field of usefulness.

Rectangles enable one to work still more freely with areas than is possible with the bar. The cube, or at least some form of solid, would seem at first thought to be somewhat more valuable than the rectangle in that it adds the third dimension. It is questionable, however, in view of the lack of skill shown by numerous makers of graphs, if to date the use of areas and solids in graphic presentation has not contributed more of confusion than of clarity.

Charts Which Confuse

The difficulty is well illustrated by this chart, prepared by the overzealous head of an accounting department, picturing his factory's production of shoes "in round numbers." (See Figure 29.) Are these three figures to be interpreted on the basis of diameter, of surface, or of cubic contents? In the first case the gain would be 1:2:4, in the second 1:4:8, in the third 18:64, certainly a considerable variation due solely to one's interpretation. It may be added that in this particular instance the comparison by diameter was intended, notwithstanding the use of the term "round numbers."

1907

О

1912

1917

Figure 29. Factory Production of Shoes Shown "in Round Numbers" In diameters these illustrations are as 1:2:4; in areas as 1:4:8; and in volumes 1:8:64. The diagrams should indicate clearly which interpretation is intended and include the data as well. In such cases it is preferable to use some forms of graphic device in which this chance for confusion does not arise.

The circle is in common use as a means of showing component parts. Public utility corporations during the last few years have frequently shown in this way the disposition of their gross revenues, an excellent choice for the purpose since no other graphic device has become so widely known nor so generally understood. In view of the erroneous theories con

cerning cost, and the relation of his wages to cost held by the average wage-earner, such charts would serve a useful purpose generally in combating the irresponsible statements of agitators.

Curves, the Graphic Device Par Excellence

The device which is most commonly employed as a means for graphic presentation is the curve. The reasons for this we may illustrate by citing the control chart recommended by Mr. Farnham which shows the total net profit accruing from a business each month in comparison with the standard set on a proper return upon the investment. (See Figure 30.)

Figure 30. Control Curves Showing Gross Earnings of Three Factories (Reproduced by permission from the Engineering Magazine.)

The standard profit, in this case ten per cent, has been represented by a straight line drawn across the chart at the proper height, and the danger point below which net profits should

not fall if adequate returns are to be earned, in this case set at five per cent, has been represented by a lower broken line. For purposes of comparison the average net profits for the previous year are shown by a third horizontal line. These three lines, one of them chiefly of historical interest and the others

Figure 31. Curves Used as a Means of Executive Statistical Control (Reproduced by permission from the Engineering Magazine.)

setting forth the results predetermined by the management, comprise what may be termed the preliminaries.

The gross earnings month by month derived from the firm's three factories are shown in Figure 31 by the three control curves marked Works No. 1, Works No. 2, and Works No. 3. Certain deductions, such as interest, discount, com

missions, and other items not easily charged to the proper factory, are subtracted from the total profits derived from the three plants and the balance is shown by the "total net profit" curve.

Variations in this total net profit curve can readily be traced to their respective causes. When profits slump badly in July and August the executive need only glance at his curves to see that the returns from Works No. I were normal and that Works No. 3 was surpassing its record of last year. Works No. 2, however, calls for attention. Were curves available showing in detail the performance of the three plants, it would be a simple matter to pass from this master chart into each further analysis as appeared desirable.

The control curves by their ability to keep clearly before the executive the essentials of the information which he needs, thus give him a real grip upon the performance of the business. Accurate, clear, thought-provoking, comprehensive in its capacity to handle data and admirable in the comparison it affords, a curve is the graphic device par excellence.

Standards for Graphic Presentation

In order to attain its highest efficiency as a language presenting data with speed and clearness, the graphic method should utilize a set of somewhat carefully standardized symbols. Unless the graphs prepared by a half-dozen men working independently upon the same data show a reasonable degree of similarity, for example, it is to be expected that much needless effort will continue to be consumed in the interpretation of diagrams whose message should be conveyed to the reader at a glance.

While considerable unnecessary variety now obtains, standards are in process of being worked out. The American Society of Mechanical Engineers some time since invited a number of other scientific societies of national scope to co