In memoriam Jan van Ettinger
In: Statistica Neerlandica, Band 35, Heft 4, S. 185-188
ISSN: 1467-9574
16 Ergebnisse
Sortierung:
In: Statistica Neerlandica, Band 35, Heft 4, S. 185-188
ISSN: 1467-9574
In: Statistica Neerlandica, Band 23, Heft 3, S. 213-225
ISSN: 1467-9574
SummaryLegal security under uncertaintyThere are two sources of losses in decision making: collective judgment and uncertainty. The situation can be summarized as follows:Minimization of losses can be achieved by using optimal decision rules. The effect of this method differs in the three cases treated. The analogies and differences of the three cases are discussed in this paper.Lezing gehouden op de Statistische Dag 1969.
In: Statistica Neerlandica, Band 18, Heft 2, S. 175-187
ISSN: 1467-9574
In: Statistica Neerlandica, Band 18, Heft 1, S. 65-73
ISSN: 1467-9574
SummaryProgramming the production of dwellings in the NetherlandsProgramming for a year the building of dwellings in the Netherlands is treated by linear programming. Restrictions are introduced for man‐power, material, private capital, subsidies and foreign currency.A "program" consists of two numbers, S (number of dwellings to be built with governmental subsidies) and V (number of dwellings to be built in the free market).The objective is discussed at length. With the returning to normal conditions in building, S‐dwellings will be subject to a higher rate of depreciation than V‐dwell‐ings.A purely quantitative objective for the program is rejected in favour of a qualitative one: maximizing the value which the dwellings built in this year will possess, say, in 1975.A simple graphical solution is shown, and the effect on the program of the industrialization of building is discussed.
In: Statistica Neerlandica, Band 13, Heft 3, S. 389-394
ISSN: 1467-9574
SummaryNormalisation as a decision problem.Mass production is the base of the enormous increase of prosperity in the western countries. Normalisation, which means a restriction of the number of types, quality etc. of a certain commodity, is an essential part of this development. However, normalisation has one disadvantage: no account can be taken of the wishes of the individual consumer. Consequently a loss‐function belongs to every imaginable set of types of a given commodity. The set of types must be chosen in such a way that the loss has its minimum value.
In: Statistica Neerlandica, Band 13, Heft 2, S. 155-169
ISSN: 1467-9574
SummaryThe article contains a short review of a campaign, undertaken in Holland during the last 12 years to arrive at standardized sizes for women's and men's clothing. The first campaign was undertaken by a department store, the second one by the clothing industry.The operational value of a sizing system depends on the value of three parameters which are not, of course, independent of each other: coverage of the system, number of sizes and average cost of alterations.The basis of the standardization is furnished by measuring a sample of the population of customers and by measuring the fitting‐tolerances of clothing.The proposed sizing systems are:a. Ladies' dresses — a two‐dimensional system based on waist girth and length of back, containing 14 sizes and giving a coverage of 90%.b. Men's suits — a system with a variable number of dimensions (2 to 4). The identification dimensions are waist girth, length of leg, hip girth and inclination of shoulder. The proposed system gives a coverage of 90% with 34 sizes. Both systems have been experimentally verified.Stress is laid on the necessity of an international sizing standardization.
In: Statistica Neerlandica, Band 10, Heft 1, S. 1-17
ISSN: 1467-9574
SummaryThis paper is an introduction to six short lectures about the theme "choice and chance", implicated in different fields of human activity. It throws some light on the history, the methods and the importance of Operations Research. Attention is paid to the elements of the way which leads from putting the problem to taking the decision. A general illustration is given by means of a simplified example. Some remarks are made about post‐war development, task and future of O.R.
In: Statistica Neerlandica, Band 9, Heft 1-2, S. 47-69
ISSN: 1467-9574
SummaryThe normal or jigsaw‐puzzle method of planning cannot be applied to the production of an engine factory. One has to fall back upon a kind of statistical automatism, like in the planning of road traffic as opposed to the planning of railway traffic.In order to do this kind of planning efficiently, it is necessary to know the statistical relationships between the degree of occupation (number of working hours divided by number of available hours) of the machines, the waiting time and the velocity of flow through the factory.The statistical analysis of some of these quantities in an Amsterdam plant, manufacturing medium size diesel engines, showed that Erlang's formulas of waiting time in telephone traffic (with exponential distribution of holding times) are applicable.These formulas are used to prove that the highest degree of occupation is not the best one from an economic point of view. Formulas and graphs are given for finding the optimum degree of occupation in engine factories and other works where the same conditions apply.
In: Statistica Neerlandica, Band 5, Heft 1-2, S. 7-16
ISSN: 1467-9574
SummaryThe establishing of standards in the transport trade by correlation analysisA firm of wholesale grocers have to deliver, once a week, goods to their clients according to their varying specifications. Delivery is effectuated by the firm's motor trucks.Observations of standing and moving times were made by means of automatically recording instruments.Standard time of delivery is given by the equation:T = 2,29G + 10,64K + 1,82A + 19,11T… standard time in minutesG… total weight to deliver, in 100 kgK… number of clients to visitA… total distance to be covered, in km)
In: Statistica Neerlandica, Band 4, Heft 3-4, S. 93-100
ISSN: 1467-9574
In: Statistica Neerlandica, Band 3, Heft 5-6, S. 224-226
ISSN: 1467-9574
Summary(The average outgoing quality limit for single sampling plans with c = o). The average outgoing quality limit occurs at P
i= 1/(1 + n) Its exact value is given by (6) and two approximations by (7) and (8), of which the more simplified formula (8) gives better results (see Table 1).
In: Statistica Neerlandica, Band 2, Heft 5-6, S. 206-227
ISSN: 1467-9574
Summary (Superposition of two frequency distributions)Notation:n: number of observationsM: arithmetic means̀: standard deviationμr: rth moment coefficientβ1: coefficient of skewnessβ2: coefficient of kurtosis.The suffixes a and b apply to the component distributions. The suffix t applies to the resulting distributions. The problem: Given the first r moments of two frequency distributions (to begin with μ0). Find the first r moments of the distribution resulting from superposition of the two components (r≥5).Formulae [1]. … [5] (§3) give the results in their most general form up to μ4.Some special cases are treated in § 4, and eight different cases of superposition of two normal distributions in § 5.In § 6 some remarks are made about the reverse situation, i.e. the splitting into two normal components of a combined frequency distribution.
In: Statistica Neerlandica, Band 1, Heft 3, S. 131-135
ISSN: 1467-9574
SummaryIn packing powders, fluids and the like in mass production there are three frequency distributions of weights, viz. net weights, gross weights and tare weights. It is shown that by elementary statistical methods the standard deviation of one of these three distributions may be found by analysing the other two.The excisting correlation, e.g. between net weights and tare weights, can be found by the same methods.
In: Statistica Neerlandica, Band 1, Heft 3, S. 156-160
ISSN: 1467-9574
In: Statistica Neerlandica, Band 1, Heft 1, S. 11-24
ISSN: 1467-9574
SummaryTolerance problems in mechanical engineering ought to be treated with statistical methods.It is shown that the use of range (instead of standard deviation) for measuring variability in tolerances and fits leads to systematic errors.The suggestion is made to measure variability in mechanical engineering by T= 4,65 s̀, being the 98%‐range in the case of normal distribution.A simple statistical approach to selective assembling is mentioned.