برآورد هزینه های پارامتریک بر اساس هزینه یابی مبتنی بر فعالیت : یک مطالعه موردی برای طراحی و توسعه بخش های چرخشی

Modern manufacturing systems are facing a globally competitive market to an extent not experienced before. This competitive pressure forces manufacturers to produce more products with shorter life span and better quality, yet at a lower cost. To succeed in this environment, manufacturing firms need to have an accurate estimate of product design and development costs. This is especially important since the shorter life span of products accentuates design and development stages. This paper presents a cost-estimation model that links activity-based costing (ABC) with parametric cost representations of the design and development phases of machined rotational parts. It also presents several parametric models applicable at design time by using parts’ feature geometry. A comparison of various parametric cost models is provided, reflecting on the accuracy of the various approaches. There is an evident trade-off between the details embedded in the parametric representation and the cost-estimation accuracy. Thus, more detailed methods, which can be used later in the life cycle of the product, are more precise, while these methods that can be used as early as design time have a higher but acceptable error rate.

Product cost estimation is a challenging task, especially during early stages of the design and development phase. This activity is crucial to the financial success of manufacturing firms and is gaining increasing importance in the contemporary competitive environment. This competitive drive forces manufacturing firms to introduce more and more products with shorter life spans and better quality, yet at an increasingly lower price. This competitive pressure propagates through the entire supply chain, requiring accurate cost estimation from all manufacturers on the chain. Traditional cost systems are known to distort the cost information by using traditional overhead allocation methods. Activity-based costing (ABC), on the other hand, has been recognized as a more accurate cost-estimation and calculation method. In our previous work (Ben-Arieh and Qian, 2003a), a modified ABC method is demonstrated and verified using sample rotational parts developed in a controlled manufacturing environment. The method is proved to be more accurate than the traditional cost estimation provided by the shop accountant. This paper extends previous research by presenting parametric cost representations for the design and development phases of rotational parts based on the ABC method. It presents several approaches towards parametric representation of cost by using parts’ feature geometry—allowing fast and accurate calculation of the cost of completing a prototype. The novelty of the approach presented is in combining ABC with the parametric cost representation of design and development activities. These activities are more nebulous and harder to quantify than regular manufacturing activities. These parametric cost-representation methods are tested on real rotational parts developed in a controlled factory environment and are compared to the detailed ABC results. All assorted cost rates in parametric cost models are established in the controlled manufacturing environment, but the method of establishing parametric cost models based on ABC could be extended to a general shop and other manufacturing processes. The paper is structured as follows: Section 2 reviews literature related to parametric cost estimation, feature-based cost estimation, and ABC systems. Section 3 briefly presents the ABC model implemented before for design and development stages in one manufacturing shop. Section 4 presents three linear parametric models using activity drivers, machining time, part volume, and number of cutters based on the ABC model. Section 5 develops a more detailed parametric model of machining time, number of setups, and number of tools, which are representing in terms of geometrical parameters known at design time. Section 6 presents a comparison of the various cost models. Section 7 provides the summary and possible future research topics.

This paper presents a methodology for integrating parametric cost estimation with ABC. Moreover, this integration is applied to cost estimations of the design and development activities of machined rotational parts in one manufacturing factory. In order to evaluate the cost of design and development phases, design activities are analyzed and modeled with IDEF0 diagrams. This representation is used in conjunction with the ABC analysis to generate the cost of the process that starts with receiving the design requirements from the customer, and continues until a prototype is produced. The analysis of design and development activities leads to the development of various forms of parametric cost estimates. The variety of parametric cost representations provides methods that are more accurate and rely on detailed information, or methods that can be used early in product development and use more approximate information. These cost equations utilize geometric parameters available during the design steps such as features’ length, diameter, and number of different features to be used. More accurate information available later is machining time property. The cost equations are generated using a sample of 15 parts produced in a controlled factory environment, which allowed complete access to time, cost, and all other types of data. These cost-estimation models can be used for web-based cost-estimation and supplier selection (Ben-Arieh and Qian, 2003b). These parametric cost equations are proved to be accurate and fast to compute. There is an evident trade-off between details embedded in the parametric representation and the cost-estimation accuracy. Thus, more detailed methods, which can be used later in the life cycle of the product, are more precise, while methods that can be used as early as design time have a higher error rate. Considering the fact that cost estimates include the overhead allocation and indirect cost of design activities, methods presented here can significantly enhance the capability of manufacturers to estimate design and development costs. The method presented here, which established various parametric cost modes based on ABC and part features, could be also extended to other kinds of parts with different manufacturing processes. Quick and accurate estimation of the whole product life-cycle cost at early design stage is still a challenging task. An integrated system combining various cost-estimation approaches could be developed for this task in the future. Technologies such as neutral network and fuzzy analysis could be used for this. Furthermore, we could apply product cost-estimation methods to support decisions in the product family design (Qian et al., 2007). These decisions include supply chain selection, price decision, and optimal platform selection in one product family.