Industrial quality and productivity with statistical methods - Productivity and statistics

It is a privilege for me to address this Discussion Meeting, sponsored jointly by the Royal Society and the Royal Statistical Society. I should say at the outset that I am an industrialist and not an academic, nor am I a statistician. Nevertheless, I did use statistics a long time ago as a young research physicist to help me in the design of experiments, and I used error theory to try to understand my results. As an industrialist, however, I have been closely concerned with quality, productivity and unit costs. Productivity, to an industrialist, is not an academic issue; it is basically a matter of life and death. It embraces not just the operational sides of manufacturing industry, but a wide spread of activities such as distribution costs and the effective use, not merely of fixed capital but, perhaps more importantly, working capital. I decided to depart from my initial notes in order to emphasize a much broader view of productivity and its importance. I changed what I had intended to say because of the dramatic increase (in early 1988) in the value of the pound against both the Deutschmark and, particularly, the dollar. You may feel this is hardly relevant to this Discussion Meeting, but this upsurge in the value of the pound had a devastating impact on some companies. Let me illustrate by reference to British Aerospace. All sales of civil aeroplanes are denominated in dollars because of the American dominance and, with the long lead times between ordering and delivering an aeroplane, the company has to anticipate the dollar exchange rate well ahead of any sales.