Anthony Aynsley Greenfield, 1931–2019 Free

Anthony Aynsley Greenfield, known to all as Tony Greenfield, was born in Chapeltown, near Sheffield, on April 26th, 1931.

He started work as a journalist, specializing in technical and scientific reporting and feature writing. The Managing Director of a large stainless steel company asked him to join the Operations Research (OR) section because

‘The OR section is full of very clever people who write reports that we (the board) don't understand. I want you to write interpretations for us.’

In the course of this work he came across several problems that he tackled himself. One was the ‘job shop problem’ about which there were many papers in academic journals. The problem was to find a solution that could be implemented as a clerical procedure. Praise from management for his solution of this and other problems encouraged him to take a degree in statistics externally from London University.

He moved in 1965 to the steel research laboratories of the British Iron and Steel Research Association in Sheffield as the statistician in the mechanical properties section, but he soon built up a department (process computing and statistics) providing a consultancy service to other departments and to laboratories in London, Wales, Teesside and Scotland. As manager of process computing and statistics he introduced the use of contemporary multivariate statistical methods to the whole industry.

In the late 1960s, with a colleague, Sean Siday, he developed an interactive statistical analysis package that had many of the features of modern packages except that it was driven by punched cards. In 1978, Sheffield University bought the package from the Iron and Steel Research Association and, 2 years later, Queen's University Belfast installed it on their system. Frequent requests for advice on the design of experiments placed a heavy burden on Tony. He responded by developing a computing procedure to facilitate design. His algorithm was published in an internal report in 1970 and in Applied Statistics in 1976. He included it in his WinDEX software package.

In the early 1970s, before small computers were available for direct connection to machines, he persuaded the directors to support the introduction of on-line computing. He installed an IBM 1800 computer and linked it through a network of cables to research plant across an area of 3 acres to monitor and control a full size billet forging machine, a 1-tonne basic oxygen steel production unit, a torsion testing machine and several other substantial research and development plants.

In 1971, a friend in Sheffield University's teaching hospital asked him to assist in a diagnostic study of thyroid conditions. This led to the publication of a paper (April 1972) in the British Medical Journal. He also delivered a paper at a medical conference on the application of statistical discriminant analysis to the diagnosis of ABO haemolytic disease in babies.

When the research laboratories were closed in 1978, Tony was made redundant. He was immediately offered a 2-year research contract in the medical faculty of Sheffield University. A multicentre study into cot deaths had been running across the UK for several years but data coding, storage, access, analysis and reporting had not been planned. Tony introduced data management and analysis systems. He was also asked to help with many projects in the medical faculty: one was to design a diagnostic chart for identification of haemophilia carriers; another for Hoffman La Roche in Basle to recommend improvements for data management of clinical trials. He assisted academic staff with tutorials for medical undergraduates. He was awarded a doctorate for his thesis on ‘The automatic design of experiments’.

Queen's University Belfast invited him to apply for the Chair of Medical Statistics: a post he took in 1980. As well as managing and presenting courses in statistics and computing for medical undergraduates and courses in statistical approaches to diagnosis for postgraduates, he was consulted widely in the medical faculty to assist with research projects leading to many publications.

He saw the need for training of medical staff in research methods and established a 2-week course that he presented each year for 5 years. This experience was useful in the 1990s when he was encouraged by the UK Minister for Science (William Waldegrave) to produce the book Research Methods: Guidance for Post-graduates, which was first published by Arnold in 1996.

Illness forced him to take early retirement at the end of 1984 but, after recuperation, he worked as a consultant and trainer with clients from manufacturing and pharmaceutical industries and universities. He developed a computing package (MetaGen) for the modelling and simulation of clinical trials.

Tony was an active member of the Royal Statistical Society (RSS), being a Vice-President in 1977–1978 and the inaugural Chair of the Quality Improvement Committee when it started in 1984. He was a Chartered Statistician, Chair of the Business and Industrial Statistics Section and of the board of the RSS Applied Statistics journal. One of his major contributions was setting up the RSS's News & Notes series. As Editor, he later revamped this as RSS News, with its fold-up meetings card. When the RSS merged with the Institute of Statisticians in 1993, Tony then handed over to Frank Duckworth, who was the Editor until RSS News was discontinued.

As a technical journalist with a Diploma in Journalism from Regent Street Polytechnic, he wrote articles for the Gas Journal, Quarry Managers Journal, Sheffield Telegraph, Sunday Express and Sunday Mirror besides the New Scientist and multiple articles in the statistics and medical press. Tony was an active partner in a European project to promote business and industrial statistics (the European Network for Business and Industrial Statistics (ENBIS), 2002–2004) during which he produced interactive software for process training based on simulators of industrial processors for teaching industrial statistics. One of his major achievements has been in translating statistical concepts into everyday parlance so that statistical thinking can benefit a wide range of people. His inspirational leadership as a communicator of statistical thinking is documented by his 1993 paper on ‘Communicating statistics’ in Series A of the Society's journal, his activities as editor of RSS News, co-authorship of The Pocket Statistician (Arnold, 1996) and Statistical Practice in Business and Industry (Wiley, 2008) and Research Methods: Guidance for Post-graduates (Arnold, 1996, with second and third editions published by Wiley in 2002 and 2016). In addition he co-authored Design and Analyse Your Experiment using Minitab (Wiley, 2007).

Tony contributed greatly to the world of technical problem solving by using statistical methods. His influence has been felt in the UK, Europe and across the globe as evidenced by being the first winner of the RSS Chambers Medal for outstanding services to the RSS (1977) and representing the RSS as a member of the UK Parliamentary and Scientific Committee (1992–1996). The RSS gave his name to the Greenfield Industrial Medal, which is awarded for contributions to the effective application of statistical methods to the manufacturing and allied industries.

He accepted the William G. Hunter award by the American Society for Quality in 2004 for his substantial contributions to statistical consulting, education for practitioners, and integration of statistics with other disciplines as well as demonstrated excellence in communication and implementing innovative applied statistical methods—considered by some to be the Nobel prize of the statistical world. He was awarded the ENBIS George Box Medal for outstanding contributions to Business and Industrial Statistics in 2009.

Tony was prominent in setting up ENBIS in 2000 and was President in 2002–2003. In company with his wife Liz and son Lance, he was one of the most regular attendees at ENBIS international conferences. He made a substantial contribution as both a Council member and a participant in the conference sessions. He set up an annual ENBIS award for the Greenfield Challenge: ‘Tell the world, outside your circle, of work you have done, and done successfully because you used statistics’.

Tony was an unusual character who stood out from others because of his wide interest in problem solving and his active pursuit of answers. He was a lateral thinker and could empathize with people from different disciplines. This made him a very valuable member of the community and an excellent example to researchers and practitioners everywhere. He was a Life Fellow of the Royal Society of Arts. He was personable and willing to discuss any issue but was also fearless in speaking his mind regardless of opposition and dissent.

Shirley Coleman