Kellie Schneider

A generic model of equipment availability under imperfect maintenance

This paper explores the impact of imperfect repair on the availability of repairable equipment. Kijimas first virtual age model is used to describe the imperfect repair process. Due to the complexity of the underlying assumptions of this model, we are unable to derive a closed-form equation for availability. Therefore, simulation modeling & analysis are used to evaluate equipment availability. Based on initial availability plots, a generic availability function is proposed. A 2/sup 3/ factorial experiment is performed to evaluate the accuracy of this model. The maximum absolute error between the simulation output, and the corresponding values of the availability function is 3.82%. This indicates that our proposed function provides a reasonable approximation of equipment availability, which simplifies meaningful analysis for the unit. Therefore, a method is defined for determining optimum equipment replacement intervals based on average cost. Next, meta-models are developed to convert equipment reliability & maintainability parameters into the coefficients of the availability model. We expand on our initial experiment using a circumscribed central composite experimental design. We evaluate the accuracy of the meta-models for the 15 experiments & 50 random experiments within the design space. For the 50 new experiments, we compare the replacement policy obtained from analysis of the meta-model to the policy obtained directly from the simulation output. The average increase in cost resulting from the sub-optimal replacement policy is only 0.10%. Therefore, we conclude that the meta-models are robust, and provide good estimates of the parameters of our proposed availability function. By doing this, we eliminate the need to perform simulation to obtain the parameters of the availability model.

Selective Maintenance Decision-Making Over Extended Planning Horizons

¿Selective maintenance¿ models determine the optimal subset of desirable maintenance actions to perform when maintenance resources are constrained. We analyse a corrective selective maintenance model that identifies which components to replace in the finitely long periods of time between missions performed by a series-parallel system. We formulate this multi-mission problem as a stochastic dynamic program, and compare the resulting optimal infinite-horizon policy to both the optimal single-mission, and two-mission policies by executing a large numerical experiment. Our results indicate that these policies rarely differ, and that when they do, the difference in long-run mission reliability is minimal, which suggests that future work should concentrate on extending results for the single-mission problem.

Social network analysis via multi-state reliability and conditional influence models

This paper incorporates multi-state reliability measures into the assessment of a social network in which influence is treated as a multi-state commodity that flows through the network. The reliability of the network is defined as the probability that at least a certain level of influence reaches an intended target. We consider an individuals influence level as a function of the influence levels received from preceding actors in the network. We define several communication functions which describe the level of influence a particular actor will pass along to other actors within the network. Illustrative examples are presented, and the network reliability under the various communication influence levels is computed using exhaustive enumeration for a small example and Monte Carlo simulation for larger, more realistic sized examples.

Fleet performance under selective maintenance

Many organizations rely on the effective use of fleets of repairable equipment that must perform sets of sequential missions. Such fleets include production equipment used in manufacturing systems, material handling equipment used in warehousing, and transportation vehicles used in distribution. Unfortunately, most organizations do not possess sufficient maintenance resources to keep the fleet in top condition at all times. Therefore, maintenance managers must make decisions regarding the allocation of the available maintenance resources. This task falls within the domain of selective maintenance, the process of identifying the subset of maintenance actions to perform from a set of desirable maintenance actions. In this paper, we consider a selective maintenance model that is used to maximize fleet reliability, the probability that all systems within a fleet successfully complete their next mission. We evaluate fleet reliability over sequential sets of missions where system maintenance is performed only during the breaks between missions sets. A combined simulation-optimization model is used to simulate the performance of the fleet over each set of missions and identify the optimal maintenance activities to perform during each break. We use this model to analyze the behavior of the stochastic process corresponding to the number of successful missions in each set.

Assessing multi-layered social networks using reliability models

In this paper we introduce the first model to determine how best to make connections amongst a multi-layer network so to ensure a specified level of reliability amongst the components (actors) in the various levels of the system. This work is broadly applicable to any relationship in ma nufacturing, business organization or social networks in which the communication structure between independent subsystems is to be determined. Our approach utilizes concepts from the fields of reliability and network optimization to provide decision makers with the first tool for measuring the tradeoff between system reliability and the degree of information sharing in a system. Using an intelligent enumeration scheme, we provide solutions and analysis for an example network. The results support the con clusion that a significant decrease in reliability accompanies even moderate increases in the requirements of participating components (actors). They also emphasize that given the level of network design flexibility allowed in our model, numerous communication paths of equivalent system reliability are available that would be difficult to obtain without mathematical modeling.