Novel cost–tolerance model based on fuzzy neural networks

Cao, Yanlong; Zhang, H; Mao, Jianhua; Yang, Jiangxin

doi:10.1243/09544054jem1789

Cited by 9 publications

(3 citation statements)

References 15 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [140], a neural network replaces the cost-tolerance function for a specific process, demonstrating better accuracy than conventional functions due to training on a large set of cost data. In [141], a neural network for combined processes is designed to estimate the cost as a function of tolerance and additional influencing factors (machine tools, machining methods, process orders, process time, cutting tools, inspection instruments, skilled operators); these are evaluated by weighted scores to get a single influence coefficient. In [142], the parameters of cost-tolerance functions for single processes are estimated from the manufacturing difficulty of the feature with respect to the process, calculated using a method based on fuzzy logic.…”

Section: Methods For Parameter Selectionmentioning

confidence: 99%

Selection of parameters in cost-tolerance functions: review and approach

Armillotta

2020

Int J Adv Manuf Technol

View full text Add to dashboard Cite

The paper deals with the cost-tolerance functions used in minimum-cost tolerance allocation on mechanical assemblies. The consistency of the cost-tolerance functions associated with different dimensions of a tolerance chain is a necessary condition for a correct optimization of tolerance values. This requires a careful selection of function parameters, in order to take into account all the variables influencing the cost, and thus to compare the costs of the different dimensions in a common scale. The cost-tolerance functions proposed in the literature are reviewed, revealing that the parameters are often selected without explicit reference to empirical data or objective rules. The development of procedures for parameter selection has been addressed in several studies, which are also reviewed along with the needed cost data published in textbooks and reports. Finally, an original approach is proposed to build cost-tolerance functions for combined processes, for which fewer results or methods are available. The parameters of a reciprocal power function, chosen for the favorable balance between accuracy and ease of use, are expressed through an empirical relationship with some design variables; these include the nominal dimensions, and the type and size of the part features involved in the allocation problem. An application example shows that the proposed procedure results in optimal tolerance values consistent with common design criteria.

show abstract

Section: Methods For Parameter Selectionmentioning

confidence: 99%

Selection of parameters in cost-tolerance functions: review and approach

Armillotta

2020

Int J Adv Manuf Technol

View full text Add to dashboard Cite

show abstract

“…The influence of processing tolerance on the corresponding processing costs has been determined using the Taguchi function and parametric equations. In the hybrid model, which is based on fuzzy neural networks, a tolerance cost model has been developed to assess the costs of processing [15]. In this model, the cost function, which can be of exponential, linear, reciprocal and combined form, is defined depending on the processing tolerance and weight coefficients of influential factors such as tools, processing methods, sequence of procedures, processing time, method of measurement, and skills of the operator.…”

Section: Literature Reviewmentioning

confidence: 99%

Model for Simulation of Life Cycle Costs at the Stage of Product Development

Todić¹,

Ćosić²,

Maksimović³

2017

Int. j. simul. model.

View full text Add to dashboard Cite

Managing the costs of a product life cycle, which is performed at the stage of its development, has a dominant influence on achieving and maintaining the competitiveness of the product in the market and the target profit as the primary production goal. Development is of crucial importance in the product life cycle, because evaluating individual solutions for the new product's conceptual and preliminary design based on the simulation of costs of all stages of its life cycle represents the way of ensuring the design requirements for excellence. This paper presents in detail the setting and development of a hybrid model of product life cycle cost management, based on fuzzy neural networks. INTRODUCTIONLife cycle cost management begins at the stage of product development. It is the stage of selection and choice of ideas and evaluation of individual solutions in the process of conceptual and preliminary design [1], i.e. product design development in accordance with the functional and other requirements of the new product on one hand, and comprehensive analysis of competitive products on the market on the other [2].Development hugely affects the costs of individual stages of product life cycle, particularly the production costs [3], because errors arising at the stage of conceptual and preliminary design can increase production costs by as much as 60 %. Therefore, all aspects of costs of production, use and recycling, including the aspects of the total cost of the product should be taken into account as early as the stage of development.In the modern environment, developing a new product emerges as a result of the work of a competent design team, usually organized on the concept of simultaneous engineering. It involves reaching a solution which is suitable for all stages of the life cycle, from development to recycling. Supporting such a design is based on the use of DFX tools intended for excellence in design [4].Instruments of strategic cost management related to the cost of the product include cost management in all stages of its life cycle by comparing the costs of the observed product with the costs of the most important competitors. Comparing the product's cost (Cost Benchmarking) with the costs of the most important competitors as the instrument of strategic and operational management enables identifying strengths and weaknesses of the own enterprise in comparison with competition, which determines directions and development strategies of addressing the identified weaknesses. LITERATURE REVIEWThere are many traditional and modern methods which can be used to assess the life cycle costs of a new or improved product in the part of the development stage that refers to the Todic, Cosic, Maksimovic, Tasic, Radakovic: Model for Simulation of Life Cycle Costs at … 109 choice and selection of ideas for a new or improved product, as well as in the assessment of variations in the product's conceptual solutions [5].Papers presenting a variety of cost assessment methods and models in the last five to six years can...

show abstract

“…Therefore, additional assumptions are needed in the regression model; in [28], regression is used to evaluate discrete points of the cost-tolerance curve, which are then linearly interpolated. As an alternative to regression, neural networks have been trained with cost-tolerance data for both single [29] and combined processes [30]; the resulting cost-tolerance models are said to allow a more accurate approximation of actual costs. This drives allocation away from the use of an explicit cost-tolerance function, as it has also been attempted using fuzzy methods [31] and the iterative solution of an equation based on process constraints [32].…”

Section: Introductionmentioning

confidence: 99%

An extended form of the reciprocal-power function for tolerance allocation

Armillotta

2022

Int J Adv Manuf Technol

View full text Add to dashboard Cite

The optimization of dimensional tolerances requires that a cost-tolerance function is evaluated consistently for all the part features involved in a given functional requirement. This is difficult because the parameters of commonly used functions are set using cost data from various sources and on possibly different scales. As an alternative, the paper proposes a revised form of one of the available cost-tolerance functions (reciprocal power), which expresses its parameters in empirical relationship with a set of design specifications on the toleranced features. These include the nominal dimension, the shape, the surface area, and the material. Following a previous study based on cost data available in literature, the values and expressions of the parameters are validated and refined using a feature-based method for the estimation of machining cost. The properties of the extended function allow to develop a simplified method for tolerance allocation that avoids the task of solving the optimization problem; it is a modified version of proportional scaling where the initial solution satisfies optimal ratios between tolerances. The discussion of the results and an application example help to justify the proposed function on grounds of correctness, convenience, and reference value.

show abstract

Novel cost–tolerance model based on fuzzy neural networks

Cited by 9 publications

References 15 publications

Selection of parameters in cost-tolerance functions: review and approach

Selection of parameters in cost-tolerance functions: review and approach

Model for Simulation of Life Cycle Costs at the Stage of Product Development

An extended form of the reciprocal-power function for tolerance allocation

Contact Info

Product

Resources

About