Method Of Imprecision Research Articles

P03 Preparing glucose infusions in neonatal intensive care: does it matter which method is used?

AimsAdministering intravenous (IV) glucose is common on the Neonatal Intensive Care Unit. Bedside preparation of glucose solutions is often necessary, usually through addition of concentrated 50% glucose to a commercially available bag. Accuracy in the glucose concentration of locally prepared bags will be influenced by a number of factors: variable overages in IV fluid bags, method of preparation and imprecision of measurement during preparation. We aimed to assess the accuracy of three different methods of preparation which had been identified through a national survey.MethodsBags of 12.5%, 15% and 25% glucose were manufactured through the addition of 50% glucose solution to commercially available bags of 10% or 20% glucose. Three bags of each concentration, were manufactured by each of the methods below:Removal of fluid from base bag prior to addition of 50% glucose, taking into account published overage.Removal of fluid from base bag prior to addition of 50% glucose, not taking into account published overage.Addition of 50% glucose, without prior removal of fluid from base bag.Three 5 mL samples were then taken from each prepared bag and sent for analysis. Glucose concentration was measured using a quantitative spectrophotometric method. As a control, three 5 mL samples were taken from three bags each of commercially available 5%, 10% and 20% glucose infusion solutions and assayed as above.ResultsA total of 81 ‘test’ samples were sent for analysis along with 27 control samples. One 20% glucose control sample was lost in transport meaning that 80 samples were analysed. The median result for each concentration and method was calculated. For method a) where the intended final glucose concentration was 12.5%, 15% and 25%, the actual concentrations obtained were 11.2%, 13.3% and 22.9% respectively. For method b) where the intended final glucose concentration was 12.5%, 15% and 25%, the actual concentrations obtained were 12.4%, 13.4% and 22.0% respectively. For method c) where the intended final glucose concentration was 12.5%, 15% and 25%, the actual concentrations obtained were 12.1%, 13.8% and 20.3% respectively. For the 5%, 10% and 20% control solutions the median reported glucose concentrations were 5.1%, 10.3% and 19.9% respectively.ConclusionsIrrespective of method used and the intended strength, the measured glucose concentration was lower than that being aimed for. In some cases, the glucose concentration was only 80% of that intended. It is not possible to conclude that one method is superior in terms of accuracy. Although it might be possible from our results to suggest the most accurate method for each concentration, this is unlikely to be predictable as manufacturers quote overages as a range rather than an absolute value. In clinical practice, preparation of a glucose solution with a lower concentration than that expected may result in prolonged hypoglycaemia with potential neurological sequelae. An alternative to bedside manufacture of glucose infusion solutions is needed. This could include pharmacy compounding of glucose strengths not commercially available or ‘piggy-backing’ of 50% glucose onto an infusion of a commercially available strength, ideally supported by a glucose load calculator.

Risk prioritization in failure mode and effects analysis under uncertainty

Failure mode and effects analysis (FMEA) is a powerful tool for identifying and assessing potential failures. The tool has become increasingly important in new product development, manufacture or engineering applications. Generally, risk assessment in FMEA is carried out by using risk priority numbers (RPNs) which can be determined by evaluating three factors: occurrence (O), severity (S) and detection (D). Due to the vagueness and uncertainty existing in the evaluating process, crisp numbers representing RPNs in the traditional FMEA might be improper or insufficient in contrast to fuzzy numbers. Currently, the fuzzy methods and linear programming method have been proposed as an effective solution for the calculations of fuzzy RPNs. However, considering the fact that fuzzy RPNs are determined on a multidimensional scale spanning O, S and D along with their interactions under a fuzzy environment, several gaps should be bridged in the evaluation, calculation, and ranking of fuzzy RPNs. First, decision makers tend to use multi-granularity linguistic term sets for expressing their assessments because of their different backgrounds and preferences. Second, numerical compensation may be existed among O, S and D that can derive different RPNs in the engineering applications. Third, the complete ranking results for fuzzy RPNs may be easily changed by the effects of uncertain factors. In this study, a fuzzy-RPNs-based method integrating weighted least square method, the method of imprecision and partial ranking method is proposed to generate more accurate fuzzy RPNs and ensure to be robust against the uncertainty. A design example of new horizontal directional drilling machine is used for illustrating the application of the proposed approach.

A fuzzy QFD program modelling approach using the method of imprecision

The use of quality function deployment (QFD) to aid decision making in product planning has gained extensive international attention, but current QFD approaches are unable to cope with complex product planning (CPP) characterized by involving multiple engineering characteristics (ECs) associated with significant uncertainty. To tackle this difficulty, in this paper, fuzzy set theory is embedded into a QFD framework and a novel fuzzy QFD program modelling approach to CPP is proposed to optimize the values of ECs by taking the design uncertainty and financial considerations into account. In the proposed methodology, fuzzy set theory is used to account for design uncertainty, and the method of imprecision (MoI) is employed to perform multiple-attribute synthesis to generate a family of synthesis strategies by varying the value of s, which indicates the different compensation levels among ECs. The proposed methodology will allow QFD practitioners to control the distribution of their development budget by presetting the value of s to determine the compensation levels among ECs. An illustrative example of the quality improvement of the design of a motor car is provided to demonstrate the application and performance of the modelling approach.

Aggregation functions for engineering design trade-offs

The choice of an aggregation function is a common problem in multi-attribute decision making (MADM) systems. The Method of Imprecision (M oI) is a formal theory for the manipulation of preliminary design information that represents preferences among design alternatives with the mathematics of fuzzy sets. The M oI formulates the preliminary design problem as a MADM problem. To date, two aggregation functions have been developed for the M oI, one representing a compensating strategy and one a non-compensating strategy. Much of the prior fuzzy sets research on aggregation functions has been inappropriate for application to engineering design. In this paper, the selection of an aggregation function for MADM schemes is discussed within the context of the M oI. The general restrictions on design-appropriate aggregation functions are outlined, and a family of functions, modeling a range of trade-off strategies, is presented. The results are illustrated with an example.

Research on application of fuzzy sets to design and manufacturing integration

Kevin N. Otto Robert N. Noyce Career Development Assistant Professor Engineering Design Research Laboratory Massachusetts Institute of Technology 77 Massachusetts Avenue, Room 3-449 Cambridge, Massachusetts 02139 December 1, 1995 The multi-disciplinary field of engineering design is undergoing an important transformation with the advent of new tools, structured product development methods and practices, and most importantly a newly developing active research community. Parametric feature-based CAD tools, genetic algorithms, and design-for-manufacture analysis methods are just a few of the recent advances spawned into practice from the design research community. Fundamental research continues in the general domain of automated and semi-automated design tools. Work by Prof. Erik Antonsson (at the California Institute of Technology) and I has included development of a structured, formal quantitative approach to design employing fuzzy sets, termed the Method of Imprecision [5, 11, 14, 16, 18, 19]. In this work, we not only developed methods and industrial case studies [10], we also clearly demonstrate that fuzzy sets are the only appropriate choice of uncertainty mathematics in many cases [17]. We also demonstrate the appropriateness of different mathematics for different uncertainty sources, from designer uncertainty in choice, environmental and production errors, and differences in the uncertain choices of downstream agents in the product life cycle [8]. It was a pleasure to note that Professor Terano, the Managing Director of Laboratory for International Fuzzy Engineering Research (LIFE) in Japan, also has now begun to note the importance of fuzzy sets in engineering design with his keynote address at the last IEEE conference on Fuzzy Systems in Yokohama [20]. Current work by my research group at MIT is focused upon integration issues in large system design and manufacturing, to which fuzzy mathematics offers a support mechanism. Many development projects today are characterized by large teams of designers and engineers who must create a large set of varied new parts and assemblies, and further must create the related manufacturing systems to fabricate and build these products. Currently there are very few formal information sharing tools for development teams, there are only informal methods such as group meetings. We have research underway to explore and develop technology and methods to allow design teams to quantitatively integrate manufacturing and design information during a product

Imprecision in Engineering Design

Methods for incorporating imprecision in engineering design decision-making are briefly reviewed and compared. A tutorial is presented on the Method of Imprecision (MoI), a formal method, based on the mathematics of fuzzy sets, for representing and manipulating imprecision in engineering design. The results of a design cost estimation example, utilizing a new informal cost specification, are presented. The MoI can provide formal information upon which to base decisions during preliminary engineering design and can facilitate set-based concurrent design.

Imprecision in Engineering Design

Methods for incorporating imprecision in engineering design decision-making are briefly reviewed and compared. A tutorial is presented on the Method of Imprecision (MoI), a formal method, based on the mathematics of fuzzy sets, for representing and manipulating imprecision in engineering design. The results of a design cost estimation example, utilizing a new informal cost specification, are presented. The MoI can provide formal information upon which to base decisions during preliminary engineering design and can facilitate set-based concurrent design.

Tuning Parameters in Engineering Design

In the design and manufacture of mechanical devices, there are parameters whose values are determined by the manufacturing process in response to errors introduced in the devices’s manufacture or operating environment. Such parameters are termed tuning parameters, and are distinct from design parameters which the designer selects values for as a part of the design process. This paper introduces tuning parameters into the design methods of: optimization, Taguchi’s method, and the method of imprecision (Wood and Antonsson, 1989). The details of the mathematical formulation, along with a design example, are presented and discussed. Including tuning parameters in the design process can result in designs that are more tolerant of variational noise.