Скачать 99.68 Kb.
The example chosen illustrates a design that is difficult to cost because of a change in process i.e. use of composites rather than metal for the manufacture of a wing. In the early study of design options, parametric cost models are often used to statistically relate cost to factors such as weight and manufacturing process. Composites have different cost drivers than metals so this invalidates the use of parametric models based on metals in order to cost composite structures and products. The same is true for new processes such as superplastic forming and high speed machining, whose characteristics are significantly different from previously used processes. Therefore parametric models based on processes used already cannot be re-applied for the new processes, as no historical information is available.
The project was to create software for the purpose of costing a product where parametric costing was not viable. The customer specified that this should be a short project to allow the costing of manufacture of a composite wing box, and that a spreadsheet must be used for this due to the availability of this package. Costing of composites is an important area of research, as designers want to make use of the strength and weight properties of composites but need to be confident that the utilisation of such components is feasible and cost effective. This spreadsheet proved to be successful.
Spreadsheet Composite Wing Box Costing System
The composite wing spreadsheet is a decision support tool for designers and manufacturers to evaluate the options for the design and manufacture of composite wing box components. It covers 4 main components, Skins, Spars, Ribs, Stringers and possible manufacturing techniques for each. The spreadsheet begins by providing the user with a page that enables the choice of component to cost. A diagram of a generic wing box and diagrams of generic components are provided to visualise the general shape and use of these components. Help pages are also provided at all stages of the costing. This start page is shown in Figure 1.
Figure 1 - Wing box spreadsheet start page
On choosing a component, in this case a spar, properties of the component are shown with default values. Colour coding is used to indicate to the user what values are editable. The derived values are recalculated to reflect any changes made by the user. When the user is satisfied with the definition of the component he or she can press the 'Define Processes' button to begin choosing and defining the manufacturing process(es) to be used, in a similar way to that used for choosing and defining the component. The interface is illustrated in Figure 2.
Figure 2 - Start of the costing of a Spar
Visual basic code was used in the spreadsheet to provide navigation and constrain the navigation order to ensure decisions were made in the correct order. However there were problems that this did not solve that will emerge in the longer term. These problems relate to maintenance, extensibility, ease of use, and sharing of information.
Problems with this approach
Maintenance is a major problem. Although the software is popular and is still being used, there are several pitfalls that will eventually end its' usefulness. Firstly although colour coding is used to indicate which fields should be edited, people inevitably make the mistake of editing other cells. If for example a user overrides a formula with a value, future cost calculations will be incorrect. In order to prevent this, code was written that protects the cells which are not editable, if the user switches off this protection and tries to edit such a cell, this triggers the cell protection to be switched back on, and prevent this editing. In the system as a whole there is more code to deal with such situations than there is to provide navigation and calculation of cost. This means there are much more lines of code than a future maintainer might expect. Even so it is impossible to envisage and prevent everything a user might do that could corrupt the calculations as Excel provides a wide range of menus and options. Therefore it would probably have been easier to begin developing software from scratch and add facilities than to use those provided by Excel and attempt to prevent their accidental misuse.
In order to provide all the options for costing of the components and manufacturing processes the spreadsheet needed to be very large. There are many sheets and thousands of fields and formulae. The information in the spreadsheet is structured by means of grouping it on the appropriate sheet in table form, but this is a structure that is hard to maintain. To ensure a valid calculation it is important to audit the spreadsheet to make sure all default values and formulae are reasonable, and correctly labelled. This is a major task and if this task is not undertaken often errors will creep in. If the components' design or manufacture changes, major changes would be needed to the spreadsheet in response, and these would be time consuming and difficult.
The maintenance problems explained above also impact on the extensibility of the spreadsheet. If this spreadsheet was to be extended or re-used to cost different components or processes, it would be very difficult to find all the relevant values and formulae to change. It would probably be easier to begin development of a new piece of software.
Ease of Use
Feedback indicates the system is reasonably easy to use because the interaction required is limited to is to editing values and pressing buttons to go forward or back. Many people are familiar with this form of navigation from using the web toolbar. However the users know they are using a spreadsheet not a web page and try to use it in generic ways that were not intended for this software. When they are prevented from exploring the alternative forms of navigation such as scrolling around a sheet or clicking on sheet tabs they may become confused.
Sharing of Information
It might become necessary to export the information from the spreadsheet to a process-planning tool for example. The lack of structure in the information makes it difficult to export it to other software systems. Because the information is a flat structure exporting it in a tree based W3C [World Wide Web Consortium] standard such as XML (eXtensible Markup Language) or RDF (Resource Description Framework) would entail the development of more customised software. There are similar problems in importing information from other systems. Knowledge sharing is essential for collaboration. [Merlo and Girard] explain the necessity for collaborative information systems for designers and the need for an object-oriented approach to this.
|Keywords Distributed decision, constraint propagation, negotiation, decision support. Introduction||Development of a Product model to support engineering change management|
|Chapter 9 e-business Decision Support Learning Objectives||Decision Support and Expert Systems (mis 538)|
|A spatial decision support system for bank location: a case study||Representing and using legal knowledge in integrated decision support systems: DataLex WorkStations|
|Decision support and disease management: a Logic Engineering approach John Fox and Richard Thomson||Support for Graphical Modelling in Bayesian Network Knowledge Engineering: a visual Tool for Domain Experts|
|Product news plan b sales double, Glamour poll shows strong support for Rx-to-otc switch, wcc welcomes new staff member||[Abel 1992] Abel, D. J., Yap, S. K., Ackland, R. G., Cameron, M. A., Smith, D. F. et Walker, F. G., Environmental decision support system project: An|