288 



National Resources Planning Board 



States Government in the purchase of munitions. 

 However, it must still be rated as a relatively undevel- 

 oped field. Of it, Captain Leslie E. Simon, Ordnance 

 Department of the United States Army, says : 



Statistical methods have proved to be a powerful tool in the 

 critical examination of some ammunition specifications prior to 

 final approval. Their use, either directly or indirectly, is almost 

 essential in determining a reasonable and economic standard of 

 quality through the method of comparing the quality desired 

 with that which can be reasonably expected under good manu- 

 facturing practice. In like manner, the statistical technique 

 renders a valuable service in framing the acceptance specifications. 

 Through its use the quantity and kind of evidence which will be 

 accepted as proof that the product will meet the standard of 

 quality can be clearly expressed in a fair, unequivocal, and opera- 

 tionally verifiable way. 



Conclusion 



It is perhaps imusual to conclude a survey of this sort 

 by stating the impressions which it has made upon its 

 writer. In the present instance, however, the element 

 of self-education has been so large that these impressions 

 may summarize the report better than any more formal 

 recapitulation. They are: 



(1) Because of its general significance as the language 

 of natural science, mathematics already pervades the 

 whole of industrial research. 



(2) Its field of usefulness is nevertheless growing, 

 partly through the development of new industries such 

 as the aircraft business, and partly through the incor- 

 poration of new scientific developments into industrial 

 research, as in the apphcation of quantum physics in 

 chemical manufacturing and statistical theory in the 

 control of manufacturing processes. 



(3) The need for professional mathematicians in 

 industry will grow as the complexity of industrial 

 research increases, though their number wUl never be 

 comparable to that of physicists or chemists. 



(4) There is a serious lack of university courses for 

 the graduate training of industrial mathematicians. 



(5) Management, which is already keenly alive to 

 the importance of mathematics, is also rapidly awaken- 

 ing to the value of mathematicians and the pecuhar 

 relationship which they bear to other scientific per- 

 sonnel. 



This last observation is not trivial. There was a 

 day when, in engineering circles, mathematicians were 

 rather contemptuously characterized as queer and in- 

 competent. That day is about over. Just now, an 

 attitude more commonly met is one of amazed pride in 

 pointing to some employee who "isn't like most mathe- 

 maticians; he gives you an answer you can use, and 

 isn't afraid to make approximations." As the proper 

 function of the industrial mathematician becomes better 

 understood, these proud remarks will no doubt cease. 

 Those who are adapted to the job will be taken for 

 granted; the others will be recognized as personnel 

 errors and not mistaken for the professional type. 

 Perhaps the present report may speed this day. If so, 

 it will have been a service to the profession and to 

 industry. 



Bibliography 



Books 



FisHEH, R. A. The design of experiments. Edinburgh, Lon- 

 don, Oliver and Boyd, 1935. 252 p. 



Shewhart, W. a. Statistical method from the viewpoint of 

 quality control. Washington, Department of Agriculture, 

 Graduate School, 1939. 155 p. 



Journal articles 



EusTis, Warner. Why the Kendall Company is interested in 

 statistical methods. Industrial Statistics Conference, M. I. T., 

 Cambridge, Mass., September 8-9, 1938. Proceedings, p. 

 143-144. 



PiCKARD, R. H. Application of statistical methods to produc- 

 tion and research in industry. Chemistry and Industry, 5!;. 

 1008 (1933). 



