Eusébio P. Nunes

Uncertainty Components in Performance Measures

Data quality is a multi-dimensional concept and this research will explore its impact in performance measurement systems (PMSs). Despite the large numbers of publications on the design of PMSs and the definition of critical success factors to develop Performance Measures (PMs), from the data user perspective there are possibilities of finding data quality problems, that may have a negative impact in decision making. This work identifies and classifies uncertainty components of PMSs, and proposes a qualitative method for PMs’ quality assessment. Fuzzy PMs are used to represent uncertainty that is present in any physical system. A method is also proposed to calculate an indicator of the compliance between a fuzzy PM and its target value, that can serve as a risk indicator for the decision-maker.

RELIABILITY ENGINEERING OF LARGE JIT PRODUCTION SYSTEMS

This paper introduces the rationale and the fundamental elements and algorithms of a reliability engineering methodology combining analytical and simulation tools, and discusses its application to the design of a large, multi-cell and heterogeneous production system with just-in-time (JIT) deliveries. In order to cope with the inherent complexity of such analysis, a two level hierarchical modeling and evaluation framework was developed. Local models are first obtained from the failure and repair processes of the manufacturing equipment. Then, these models are combined with the failure propagation delays introduced by the work-in-process buffers in order to obtain the system level model. The evaluation algorithm is able to deal with reliability models containing stochastic processes with generalized distributions. This fundamental requirement comes from the fact that repair and failure propagation processes typically present hyper-exponential distributions, e.g., lognormal distributions, that cannot be assessed using the conventional reliability techniques. The second part of the paper addresses several design issues of the production system that directly impact the reliability of the deliveries, and explains how the behavioral and structural characteristics of JIT production systems were explored in order to implement effective evaluation algorithms.

Developing organisational learning through QC story

Continuous improvement or more generally improvement is a necessity for companies in a dynamic context. The effectiveness and efficiency of such improvement can foster company sustainability, particularly if it exhibits characteristics of a learning organisation. This paper shows a methodology to develop organisational learning through the use of one quality improvement methodology known as Quality Control (QC) Story. The work is based on a case study from a manufacturer of electronic products. It describes QC Story application over 10 years and its effects on key performance indicators. Organisational learning is assessed through a questionnaire survey. A method to manage knowledge based on a lessons-learned procedure is described and a generalisation is induced. Results suggest that repetitive use of QC story along with the presented procedure of lessons-learned contributes to improving processes performance over long periods. It is also concluded that this company using the presented procedure exhibit characteristics of a learning organisation.

Evolution of process capability in a manufacturing process

Before mass production of individual components industries may assess its capability to produce it according to specifications. This capability assessment is a common requirement in the automotive ...

Statistical process control automation in the final inspection process: An industrial case study

This case study arises from the need to make more robust and effective quality assurance procedures of the products by automating the final inspection process. The case study explains how the automation of the inspection process was performed in a company from the automotive sector. Knowledge, involvement and commitment of operators and respective managers should not be neglected because their reaction against the change procedures influence the success of any automation performed. The successful introduction of automation contributed to a more efficient process and from the pilot station to the remaining stations problem solving and continuous improvement was evidenced.

On the estimation of hospital beds occupancy after hip surgery

For hospitals, the variability on demand (number of daily urgent admissions) and the variability in the Length of Stay (LoS) (bed occupancy) may affect the quality of service provided to patients and the effectiveness of the overall service. This paper studies the LoS of 238 patients who performed hip surgery in the orthopedic service of a Portuguese hospital in 2014. It uses variables available in electronic databases, such as Age, Gender, ASA classification; Surgical Apgar Score, Type of hip surgery; Weekday of the surgery; Starting hour of the surgery and Duration of surgery to predict LoS and provides a model that correctly indicate if a patient stays more than 7 days in 72.1% of the cases.

Methodology for uncertainty characterization of performance measures

Purpose The purpose of this paper is to present a methodology to represent the uncertainty generated in performance measures (PMs) during the operational step or “use step” of the performance measurement process (PMP). The different steps of the methodology are described and exemplified through an application example. Design/methodology/approach An index that reflects the level of uncertainty originated by the factors and the strength of their inter-relationships is developed through the use of graph theory. A graph is developed considering all sources or factors of uncertainty that may be present in this process. Based on the graph, the methodology includes the use of a matrix and the determination of the associated permanent function which is used for determining the uncertainty index. Findings During the development of the methodology, it was found that the use of a scale that includes zero for assigning values of the elements of the matrix is not appropriate when using graph theory and permanent function calculation, since in this case the permanent function is insensible to changes in some matrix elements. Originality/value This paper identifies all the sources that can affect the quality of performance measurement values during the operational step of the PMP and proposes a method to characterize the strength of this uncertainty. Beyond alerting decision makers to the level of uncertainty associated with a PM, it also allows defining appropriate actions to improve PMs’ quality.

A Fuzzy Logic Approach in the Definition of Risk Acceptance Boundaries in Occupational Safety and Health

Organizations need to make decisions about risk acceptance, in order to decide about the need of risk-reducing measures. In this process, the personal judgments of Occupational Safety & Health (OSH) practitioners have a great importance. If on one hand they have the technical knowledge about risk, on the other hand the decisions can be dependent on their level of risk acceptance. This paper analyzes judgments of OSH practitioners about the level of risk acceptance, using Fuzzy logic approach. A questionnaire to analyze the reported level of risk acceptance was applied. The questionnaire included 79 risk scenarios, each accounted for the frequency of an accident with more lost workdays than a given magnitude. Through the two-step cluster analysis three groups of OSH practitioners were identified: Unacceptable, Tolerable and Realistic groups. A further analysis of the realistic group judgments about risk was performed, using the Fuzzy logic approach. The fuzzy sets of inputs and output variables were determined and the relationship between the variables was mapped through fuzzy rules. After that, the Min–Max fuzzy inference method was used. The obtained results show that the risk level is acceptable when input variables are at the lowest value and unacceptable when the risk level is high. The obtained results allow to better understand the modeling of OSH practitioners’ judgments about risk acceptance, being noted the uncertainty related to these judgments.

Uncertainty characterization of performance measure: a fuzzy logic approach

The process of performance measurement encompasses the activities of designing, data collection and analysis. The lack of quality of performance measures (PMs) may influence decision-making. Since the process of performance measurement involves generally several actors, the decision-maker may not be aware of the level of uncertainty associated with performance measures. In this paper, fuzzy logic is used to represent the uncertainty generated in PMs during its design, use and analysis stages. The identification of uncertainty sources and the determination of an Uncertainty Index support actions to improve performance measures’ quality. An application example is provided to show the usefulness of the proposed methodology.

The Application of a Fuzzy Approach to the Analysis of OSH Practitioners Level of Risk Acceptance

OSH who accepted the challenge to participate in this study. The authors are grateful to national funds by FCT –Fundacao para a Ciencia e Tecnologia in the scope of the project: PEst-OE/EEI/UI0319/2014