Zdenek Vintr

Optimization of reliability requirements from manufacturer's point of view

Nearly all consumer and industrial products are sold with warranty. This has important cost implications for the manufacturer because every warranty presents additional costs for him. The article analyses dependencies among cost price, warranty cost and product reliability. According to these analyses, the article presents two methods of optimisation requirements on the reliability of products. The first method makes it possible to specify the requirements on reliability level, which will lead to the minimal manufacturers costs (research, development, production and warranty costs) if the length of the warranty period is firmly given. The second method makes it possible to specify the requirements on reliability level, which will lead to the maximal possible length of warranty period, if the manufacturers costs are firmly given.

Estimate of Warranty Costs Based on Research of the Customer's Behavior

This paper deals with an estimate of costs when two-dimensional warranty is granted. The paper includes a procedure of estimation of costs of the warranty based on a research of behavior of the customers. This procedure is applicable for an initial estimate of warranty costs or analysis regarding the extent of warranty to be provided. The practical use of the proposed method is demonstrated by an example of the estimates of warranty costs of a lower medium class personal car manufactured in Czech Republic. For this paper, several hundred users of the said car type were interviewed to identify the average mileage by a typical car user during one year of operations. The result of survey were assessed statistically and used to analyze the an influence of changes of basic parameters of the warranty, namely guaranteed calendar time of use and guaranteed time of operation, on the warranty costs.

Modeling and Analysis of the Reliability of Systems with One-shot Items

This article deals with modelling and analysis of the reliability of complex systems that use one-shot items during their operation. It includes an analysis of the impact of the reliability of used one-shot items on the resulting reliability of the system as a whole. Practical application of theoretical knowledge is demonstrated on an example of a model of reliability of an aircraft gun that was used for optimization of the guns design during its development and designing. The analyzed gun uses two types of one-shot items - rounds intended for the conduct of fire and special pyrotechnic cartridges designed for re-charging a gun after a failure of the round. A proposed model enables us to analyze the responsiveness of the whole system due to a change of reliability of rounds and pyrotechnic cartridges and a change of the number of used rounds and pyrotechnic cartridges. This article demonstrates utilization of a proposed model to determine an optimum number of pyrotechnic cartridges that will achieve a required probability of fulfilment of mission with the given reliability of individual subsystems of the gun, rounds and pyrotechnic cartridges. Proposed procedures of modelling and analysis of the reliability of complex systems with one-shot items were used in practice in designing and development of the PL-20 aircraft gun. Development of the gun was successfully finished in 2005, and the gun was fielded in the armament of the Czech Republic Air Force.

R&M requirements allocation in upgrading a system

The article deals with a method of R&M requirements allocation which is applicable in upgrading technical systems. The method is based on the presumption that the basic R&M parameters of main sub-systems of original design are available and the requirement for the availability level of the upgraded system is stated. The main aim of the method is a requirements allocation which ensures an increase in system availability to a required level with minimal design changes. The basis of the method is a mathematical model describing dependencies between the overall system availability and availability of separate sub-systems. The model allows testing a system-availability response to changes of sub-systems availability. By using the model, a procedure was devised for specifying the sub-system (or sub-systems) which is the most suitable for upgrading. At the same time, the procedure allows allocation of R&M requirements for this optimal sub-system (or sub-systems) so that overall system-availability will meet the stated requirements. The method was developed and used during the upgrading of a heavy tracked vehicle.

A Tool for Decision Making in K-Out-of-N System Maintenance

The paper deals with possibilities of maintenance optimisation of k-out-of-n systems. The first part of the paper briefly analyses influence of preventive maintenance concept on reliability of k-out-of-n system and presents possibilities of preventive maintenance optimisation with respect to maintenance costs. The second part of the paper presents a suggested method of maintenance optimisation in the case of a processing machine for dosing tiny loose material and smaller piece goods. The method takes into account not only maintenance costs but also production losses caused by a prospective failure of the system. The proposed method facilitates decisions on maintenance intervals modification and the range of maintenance from the economic point of view.

Does appropriate software support for target identification exist

The paper deals with the evaluation and the selection of available software tools that can be utilized in solving problems related to the identification of the targets of malicious acts. The paper presents grant project requirements that focus on the development of methods for the target identification in the field of nuclear energy and critical infrastructure in the Czech Republic. The main requirements are related primarily to the modeling of unacceptable effects of malicious acts. Based on the project purposes, the specific requirements for software support for targets identification are defined. In the next step, the available methods and techniques such as event trees, fault trees and attack trees are introduced. After that there is a summary of available software tools focused on modeling and solving event, fault and attack trees. Individual software tools made by reputable producers are characterized in detail. Finally the paper compares the capability of available software tools with the specified requirements.

FMEA used in assessing warranty costs

This article presents a proposal of procedure that enhances standard failure mode and defect analysis (FMEA) and at each mode of failure it analyses other information that are necessary for estimation of warranty costs. In addition to standard analyzed information, at each failure mode it further evaluates an assumed number of a given type of failure during a warranty period, method of its repair, its labor consumption and costs needed for spare parts and material. It also presents a model that based on these data enables to carry out an estimate of warranty costs of a product as a whole.

Evaluation of physical protection system effectiveness

The ability of physical protection system (PPS) to withstand a possible attack and prevent an adversary from achieving his objectives is generally characterized as PPS effectiveness. The evaluation of effectiveness serves then as a basis for assessing whether a relevant PPS meets an intended aim. The article presents the selected results of an extensive analysis tackling the present state of PPS system effectiveness evaluation. There has been determined the achieved state of the development of a PPS efficiency evaluation theory, and also the possibilities of practical application of known algorithms, analyses, models, program products, tests and exercises which might be used for evaluating the PPS effectiveness. The authors of the paper have proposed that a couple of ways can be selected to develop these methods more when following the results of the performed analysis.

Threat characterization in vital area identification process

The identification of a potential attackers targets is fundamental in the design of a physical protection system for complex technological facilities such as nuclear facilities. An attack can be carried out to sabotage, or gain control of, targets that are physically represented by different parts of systems, structures and components (SSC), or their combination. Those SSCs that need to be protected in order to prevent radiological consequences in a nuclear facility are called vital and are located within vital areas. The use of attack trees in the vital area identification is an innovative approach that enables modeling of both the SSCs and the threat including their mutual interaction. Threat characteristics are described by a well-designed and scaled set of attributes. Required attackers capabilities to disable each SSC have to be assessed. Methods of expert opinion elicitation are mostly used for the assessment of attackers abilities to perform malevolent events on an individual SSC represented by tree leaf nodes. This paper presents the selected findings from the attack tree implementation in the vital area identification focusing on threat and supplementary attributes for the attack tree model. The protection tree is used to evaluate and identify vital area candidates. The suggested method of threat characterization in an attack tree enables time-effective, more precise and well-documented vital area identification.

Contribution to Rolling Bearing Reliability Modeling

Rolling bearings are usually considered to be non-repaired items the reliability of which is characterized by mean time to failure, or so called basic rating life. Reliability describes these parameters well in case the bearings are used in operation up to the very time the failure occurs, or during the time corresponding with basic rating life. In case of railway applications the bearings are often used in large groups and are preventively replaced after much shorter operating time as compared with their basic rating life. In the article there is a model which enables us to describe the bearings reliability in this specific case and to specify a number of failures which might be expected from a group of bearings during operating time, or to determine mean operating time between failures of bearings.