Operations Research: The early period (1955 – 1964)
Around the time of WW II, a new research field called Operations Research (OR) surfaced. Operations Research makes use of advanced mathematics to optimise decision-making. OR was first developed as a response to the military needs in WW II. Initially, it was sparsely used for industrial and governmental activities. In 1950, Prof George E. Kimball and physicist Phillip M. Morse laid the groundwork for Operations Research in their book ‘Methods of Operations Research’, in which they give the definition: “Operations research is a scientific method of providing executive departments with a quantitative basis for decisions regarding the operations under their control”.
An important aspect of the definition is the term ‘quantitative’. Certain aspects of practically every operation can be measured and compared quantitatively with other, similar operations, and those aspects can be studied scientifically. The phrase ‘basis for decisions’ implies that these quantitative results are not the only measure used to make decisions. Other aspects may influence the decision-making as well such as politics, tradition etc.
Despite the fast and positive results Operations Research had during WW II, it took a while before TNO started with Operations Research. In 1955, Director J. L. van Soest started a task force working on optimising communication channels. Transmission and receive had to be maximised while noise had to be minimised. The task force familiarised themselves with the mathematical knowledge needed and extensively studied game theory. The task force quickly concluded that game theory was not only applicable to games like poker but could also be very useful for military applications such as optimising torpedo targeting.
However, as the WW II origin of Operations Research lay in radar applications, it seemed only natural to incorporate Operations Research in the ongoing TNO radar and fire-control systems research.
But, after a discussion in 1956 with ir Maarten van Batenburg – a PhD student at the Delft University of Technology- the Royal Army commissioned a first assignment for the task force: optimise the sonar search procedure for the ‘searchlight’ sonars PAE-1 and CWE-1 installed on the Royal Netherlands Navy Holland/Friesland class submarine hunters. After many hours of work, the task force literally wrote down every possible combination of variables. Because of this very cumbersome task and the ever-increasing mathematical complexities present for the possibilities, the idea of a simulation was considered. Nowadays a simulation seems a base Operations Research method, but in 1956 it was quite difficult to perform.
The task force simply started out with rolling dice. Later, they started using pseudo-random numbers although this was a cumbersome manual process. But it was successful. After two months of work, 45 ASW simulations were completed. The conclusion was that the differences between the different search strategies were negligible, which put quite an anti-climax on the research.
The unbeatable tic-tac-toe machine
In 1957, ir E. W. Gröneveld became the task force leader. He decided to have a tic-tac-toe machine built as a demonstrator of the zero-sum game theory. The electronic tic-tac-toe machine could never lose a game. Nevertheless, many military visitors as well as HRH Prince Bernhard tried their luck on the machine.
Unfortunately, the hard-wired min-max algorithm hardware was lost in the 1970s. The 1957 user interface still exists in the museum.
In 2023, the museum tries to bring the user interface into life again using an Arduino as processor.
The eighteen, 65 years old, lamps all work.
Between 1956 and 1964, the task force solely worked for the Netherlands Royal Navy. Besides sonar and fire-control systems, a lot of attention went to optimising the clearing of sea mines. Given known physical limitations, an optimal mine sweeping plan was developed. A lot of probability theory was involved due to the built-in activation times the mines possessed. In 1964, the first assignment from the Royal Netherlands Air Force concerned the evaluation of the 40 mm anti-aircraft (AA) artillery. The Royal Netherlands Army followed with a support request concerning the choice of a new transport vehicle.
[With thanks to T. Nooijen for providing the background on Operations Research developments during the years 1947 – 1977 in his publication]