3.1.2 ContextsThe contexts specify

3.1.2 Contexts
The contexts specify defined entities in three levels in our model. The first is defined in a process level, the second on a material level and the last is directly related to the feature. This context is a cross between a process, a material and a feature, connected to an environment (cf. Fig.2).
3.2 Sand casting modelling

Based on this analysis we created a generic model using Cost Advantage Software. The first step closely defines the production process dedicated to this industry. The master parameters acting upon the product cost must be identified in order to enrich the cost semantic of the model.

The generic approach quickly highlights the problems of the contexts characteristics. How to define a significant cost for the part and with which level of detail in order to be generic? We will not answer this question, but we will present the paths or solutions we used.

First, let us describe the sand casting organisation through the part life cycle in the enterprise. This model is based on the SMC Colombier Fontaine foundry (France), from AFE Metal group. We will reduce our cost model to the primary parts through its life cycle in the enterprise and focus on the production phase, from the sand elaboration, the tooling machining and the parts perfecting.

3.3 Step induced by the use of Cost Advantage

Figure 3 presents a functional view of the process with the compound (raw material, tooling) and the elements needed to manufacture a part linked with the major indicators dealing with the final cost (loss, scrap ratio, production rate).

Figure 4 represents a transposition under the concepts of Cost Advantage of this model gathering the three levels of entities defined in the software. To illustrate this, for example, the mould, the tooling and the cores are components required to carry out the assembly named moulding by the operation (feature) of remould. It is thus necessary to define the final part, to carry out the two assemblies, which are the moulding then the casting.

In term of model design, the functional view identifies the assemblies needed, it is then necessary to define the components and choose and define the related operations. An ascending step must be practiced, starting with the components up to the definition of the assemblies. The costs calculation is presented in figure 5 and the structure of the implemented data in figure 6.
Calculations are simply taking into account volumes of material, rates of production, losses and the machine and labour costs. To say nothing about the difficulty in knowing the exact parameters, the problems we met were model organization, more than process modelling.

The rules of calculation then implemented will make possible to the future user to inform only the relevant data about its study. Indeed, only the operational process, rates, dimensions, numbers of cores (etc.) will be required (or deduced directly in a CAD software) to allow an automatic calculation of the cost of the part according to its particular characteristics.

4 Discussion

The principal remark on this model, implemented with Cost Advantage platform, comes to that it doesn’t integrate the global aspect of the costs management. For instance, the indirect share due to the development and the design of the tools (master pattern, cores boxes…) isn’t taken in account.

But let us keep in mind the framework of use of such a tool: to help the designer to achieve a cost objective related to expected technical functions. He can then propose design modifications (joint section, cores, quality…) or process (number of parts in the mould…) to achieve its goal. Moreover, when the whole partners validate this design, the parameters costs are fixed (in agreement or not with the objective laid down initially), the cost is well known and will not be any more a consequence of later decisions.

During this work we have identified a principal difficulty, which for this modelling impact the multiplicity of the elements on the availability for their characterization and for their organization. Even if the manufacturing sand casting process seems simple, it uses many components (alloy, cores, mould…), and we limited the definition in term of model refinement since each one of these components could be the subject of a finer modelling. We tried to choose this limitation in order to represent the general process without going closer into enterprise specificities. For example, the cores or fusion process of realization are not completely defined since depending on the machines, uses and other specificities of the workshop. In fact this remark is shareable for all the components entering into the realization of the finished part. But we think that we defined a basic minimal skeleton, transposable from one company to another using the sand casting process.
5 Conclusion

To conclude on this work, we started to apprehend logic of cost oriented modelling through a tool using the concept of cost entity. In order to ensure a generic aspect of our work, we deliberately limited the details of the operations, components and assemblies. Indeed, the development of these elements takes into account many parameters that it seemed to us initially overflowing to define.

We thus defined a structure that we think minimal, as well as indicators necessary to evaluate the whole costs without the indirect part. The two next steps of this study are in the first hand among other sand casting companies, to apply this modelling, configure the model with the existing processes and informing the exact values of the indicators. But also in the other hand, to calibrate the model and the results on real studies already done. These two last steps allow comparing the effectiveness of the various companies and could be used as Benchmark. A foreseeable difficulty is the possibility of reaching this information. Moreover, these factors are often managed in a total cost accounting. As a result, the efficient indicators may be sunk in a not very transparent accounting system or may be aggregated with not relevant others.

The other significant continuation to give to this work is the taking into account of the global costs mainly related to the indirect shares (structural). Our introduction puts forward the lack of management of these aspects and our first approach did not give place to a better control of these factors. However the work is done and the workers must be paid (designer, maintenance, buyers, logistics…) even if their work is not as well managed through a cost management system. A better specification (by the means of indicators, of metric) of the tools design phases, tools lifespan, etc. could integrate a real cost of the complete series. The question of the relevance of the tool used for this type of approach arises then. Some solutions come from the use of single or very limited number of cost inductors such as the time and they define a minimal global enterprise cost per hour to balance its financial objectives. Such an approach allow a multi-level management of the parts impact and give real time information to asses the enterprise objectives and manage strategies and operational decisions.