WO2005083567A2

WO2005083567A2 - Method of testing the development status of a developed system

Info

Publication number: WO2005083567A2
Application number: PCT/EP2005/001558
Authority: WO
Inventors: Michael Schneider; Rolf Schwedhelm
Original assignee: Eastman Kodak Company
Priority date: 2004-02-26
Filing date: 2005-02-16
Publication date: 2005-09-09
Also published as: DE102004009227A1; WO2005083567A3

Abstract

The present invention relates to a method of testing the development status, in particular, the degree of quality and/or reliability, preferably the launch readiness, of a developed system, specifically a machine. An object of the invention is to provide a method which permits the differentiated testing of a system. This object is achieved in that problems (Pi), which potentially occur during the operation of the system and which are to be used for testing, are detected and/or defined, that at least one quality criterion (Kn) is used as the default for analyzing the problems, and that the degree of severity (Si) of a specific problem (Pi) is expressed in terms of the ratio of the extent of the occurrence of at least one quality criterion (Kn) regarding this defined problem (Pi) relative to the maximum possible or expected extent (Max Kn) of this specific problem (Pi).

Description

Method of Testing the Development Status of a Developed System

The present invention relates to a method of testing the development status, in particular, the degree of quality and/or reliability, preferably the launch readiness, of a developed system, specifically a machine.

In principle, a method of the aforementioned type has been known from US-A-6 038 517. However, in this method, an error-free running time is used as the general testing criterion.

An object of the invention is to provide a method which permits the differentiated testing of a system, in particular, by selecting points of special emphasis applicable to continued developments or to improvements of the system.

In accordance with the present invention, this object is achieved in that problems (Pj), which potentially occur during the operation of the system and which are to be used for testing, are detected and/or defined, in that at least one quality criterion (K_n) is used as the default for analyzing the problems, and in that the degree of severity (Si) of a specific problem (Pj) is expressed in terms of the ratio of the extent of the occurrence of at least one quality criterion (K_n) regarding this defined problem (Pj) with respect to the maximum possible or expected extent (Max K_n) of this specific problem (Pi).

In so doing, different quality criteria may be viewed as being of different importance or significance in the process of testing the system. Therefore, one development of the invention provides that the aforementioned ratio (K_n) / (Max K_n) assigned to a specific problem (Pj) for determining the degree of severity (Sj) is multiplied with a weighting factor (g_n), where 0 < g_n < 1, that is assigned to this quality criterion (K_π). If the weighting factor g_n = 0, the respectively weighted quality criterion, of course, becomes unnecessary as such. Hence, this way of weighting makes sense only when a plurality of quality criteria, as are preferably provided, are selected in such a manner that, in order to determine the degree of severity (Sj), several stated ratios (K_n)/Max K_n) or (g_n Kn)/(Max K_n) for different quality criteria among a number (n) of quality criteria Ki through K_π are multiplied with each other.

For example, in particular the following may be considered as quality criteria: the downtime of the system when a specific problem occurs, and/or the frequency of the occurrence of a specific problem, and/or the number of systems in which a specific problem occurs.

Preferably, the method of the present invention is designed and suitable for testing a digital printing machine.

Consequently, in accordance with the present invention, the degree of severity (Sj) is defined in terms of the following formula: gι * Kι * - * g_n * K„ S, MaxK_x • ... • MaxK_n

Considering this, there is a quantity of 1 through n quality criteria Ki through K_n, from among which a few or every criteria are selected for the aforementioned formula as the criteria that are significant regarding the analysis of the system, and that these quality criteria are weighted with their respectively assigned weighting factors gi through g_n. Each weighting factor has a real value of from 0 to 1. The respective maximum (bad) value for the quality criterion Max K_n is used to qualify or standardize the value Sj. A corresponding quality value can be defined in terms of Qj = 1 - Sj. Consequently, the quality value would be equal to zero, assuming that all weighting factors are equal to one, and assuming that all quality criteria have satisfied the "maximum bad' status, i.e., each problem Pj, taking into consideration each quality criterion, exhibits maximum critical status. In light of the quality criterion K_n, the degree of severity Sj and the quality value Qj must be determined for each and every problem. Thus, the degree of severity S and the quality value Q are quasi-functions of Pj and have the respective function values S (Pj) = Sj and Q (Pj) = Qj, where the index quantity I = { i | i e N }, with N, in this case, representing the finite quantity of natural numbers which are the numbers of consecutively numbered problems Pj.

Examples of inventive tests, which may result in additional inventive features, which, however, do not restrict the scope of this invention, are shown by the drawings.

They show in

Fig. 1 data and a quality graph depicting the degree of severity S, taking into consideration two quality criteria; and,

Fig. 2 a quantity graph depicting the degree of severity S, as in Fig. 1 , and the graph of the respectively assigned quality value Q.

In Fig. 1 , problems 1 through 28, i.e., Pj of Pi through P₂₈, are plotted on the abscissa. For example, considering an electrophotographically operating digital printing machine that is being analyzed, the following may be potential problems: no fusion roller temperature control, machine stop, wrong printing sheet length, paper jam, error message from a registration sensor, hood not closed, no printing material, and so on. Considering each of these problems, the right ordinate of Fig. 1 indicates how many times the respective problem has occurred, for example, in field testing. These frequency values are connected with each other by a solid line that forms a graph. Consequently, this frequency represents a first quality criterion, i.e., Ki, in testing. Using bars corresponding to the dimensions of the left ordinate of Fig. 1 , the downtimes resulting from the respective problem are depicted. In the course of testing, these downtimes represent the second significant quality criterion K₂. Of course, this kind of field testing may be performed with a plurality of systems, i.e., for example, the same type of printing machines, so that the number of tested machines displaying a given problem can be used as a third quality criterion K₃. Hence, the index n of quality criteria would run from 1 through 3.

In accordance with the aforementioned formula, the degree of severity Sj for each problem Pj would be determined mathematically as follows:

_ Downtime _ because _ P_t • Frequency _of _P_t • Machines _ with _ P_t Max(Downtime _ because _ P_l ) • Max(Frequency _of_P_l) » Number _ all _ Machines

In Fig. 1 , which was created for only one machine, it is assumed that the last factor in the numerator and in denominator, respectively, is equal to 1. In this case, the two remaining quality criteria K-i and K₂ are used for each problem Pj to determine the respective degree of severity Sj in accordance with the above formula. In Fig. 1 , these degrees of severity Sj are plotted (qualitatively) and connected by a dashed line to form a graph. To achieve this, the problems have already been sorted as to degree of severity in Fig. 1 in such a manner that the most severe problems are located on the extreme left.

Fig. 2 shows the graph depicting the degree of severity, again quantitatively, in correspondence with the dimensions of the left ordinate in Fig. 2. In addition, the corresponding graph of quality values Qj has been defined and plotted in Fig. 2.

Testing in accordance with the invention may be performed in view of the following objectives, for example:

1. If the degree of severity is below a default limit or if the quality value is above a default limit, this number can be used as a criterion to release the analyzed system for launching.

2. If the degree of severity or the quality value is based on field-testing data, these numbers may be used for prioritizing development tasks. If the degree of severity or the quality value is based on field-testing data, these numbers may be used to communicate an improvement to the customer (e.g., in the course of product iteration or after providing update kits).

Claims

Patent Claims

1. Method of testing the development status, in particular, the degree of quality and/or reliability, preferably the launch readiness, of a developed system, specifically a machine, characterized in that problems (Pj), which potentially occur during the operation of the system and which are to be used for testing, are detected and/or defined, at least one quality criterion (K_n) is used as the default for analyzing the problems, and the degree of severity (Sj) of a specific problem (Pj) is expressed in terms of the ratio of the extent of the occurrence of at least one quality criterion (K_n) regarding this defined problem (Pj) with respect to the maximum possible or expected extent (Max K_n) of this specific problem (Pj).

2. Method as in Claim 1 , characterized in that said ratio (K_n ) / (Max K_n) for a specific problem (Pj) is multiplied with a weighting factor g_n, where 0 < g_n < 1 , that is assigned to this quality criterion (K_n), in order to determine the degree of severity (Sj).

3. Method as in Claim 1 or 2, characterized in that, in order to determine the degree of severity (Sj), several stated ratios (K_n)/Max K_n) or (g_n K_n)/(Max K_n) for different quality criteria from among a number (n) of quality criteria (Ki through K_n) are multiplied with each other.

4. Method as in one of the previous Claims, characterized in that one of the quality criteria used is the downtime of the system when a specific problem occurs.

5. Method as in one of the previous Claims, characterized in that one of the quality criteria used is the frequency of the occurrence of a specific problem.

6. Method as in one of the previous Claims, characterized in that one of the quality criteria used is the number of systems in which a specific problem occurs.

7. Method as in one of the previous Claims, characterized in that the system is a digital printing machine.