Summary by Aarti Nirgudka
Master of Accountancy Program
University of South Florida, Summer 2002
Today, most businesses are moving from a narrow functional (stovepipe) view of the organization to a cross-functional (holistic) view of the organization. Moreover, models with a functional focus, still exist in many organizations and conflict with the current organizational focus of continuous improvement and being "world class." Thus, process model integration, an organization-wide view, which integrates existing models within different functions, has become the focal point of many organizations. Model integration is consistent with activity-based management and not the function-specific approach that currently prevails.
Need for Model Integration to Support Cross-Functional Analysis
Traditional functional models tend to achieve functional subsystem optimization at the expense of the overall system. Additionally, compensation systems in organizations have rewarded functional optimization and thus, reinforced a functional focus. However, in an environment characterized by continuous improvement, integrated processes seem to be of more relevance rather than the very limited functional models. A graphical illustration of the move towards model integration is presented below:
A cross-functional, process orientation implies a move up the pyramid and a change in the managerial focus. The cross-functional focus emphasizes model integration and thus, is able to provide managers with coherent information for strategic planning and control.
An Example of Model Integration
A company has the following models developed independently of each other (thus, having a functional focus):
Econometric marketing model- forecasts demand in terms of sales volume for a product for the next fiscal year,
Event simulation manufacturing model- estimates the required expense to produce enough of the product to meet the specified demand,
Transportation model- determines the minimal cost of distribution the product to customers,
Pricing model- calculates a price for a produce given a demand volume, and manufacturing and distribution expenses, and
Financial model- determines the revenues and net income from sales of the product given demand volume, manufacturing and distribution expenses, and product price.
However, these currently existing models, built in isolation of each other, are unable to provide management with useful answers to questions such as:
What effect will developing islands of automation in the manufacturing process have on net income?
What will happen to revenues if demand for our product softens as a result of decreased spending by Department of Defense?
What baseline price do we need in order to provide for customer needs while maintaining long run profit goals?
Thus, by integrating these models where the outputs of one serve as the inputs of another, the company is able to achieve global optimization instead of local optimization. An integrated example of the above models is presented below.
Nevertheless, to develop such a highly integrated model in a stovepipe organization presents many obstacles including:
Conflicting objectives of the respective models,
Availability of accurate and valid data,
Relative validity of the models, and
Software and hardware incompatibility of model implementations.
Moreover, the key to model integration is data integration. An organization must be able to develop a shared data management philosophy in order to support cross-functional model development.
The Integrated Modeling Environment (IME)
IME is a quick and easy way to support the integration of existing models developed within different functions. IME supports the overall life cycle for specific model development and also facilitates the composition of models from two or more existing, constituent model components. The latter process is known as model integration. Model integration, however, requires a shared view of the strategic goals of an organization, the business processes necessary to realize those goals, and the mathematical models required to support those business processes. This, in turn, requires an information system infrastructure that promotes the sharing of data and the models, which use the data. IME is the component of this infrastructure, which supports the development, integration, and maintenance of mathematical models compatible with an organization’s goals.
Further, a necessary requirement for model integration is that all models be expressed using a single, consistent representation ("schemata" in IME terms). Representations come in many forms like mathematical, graphical, narrative, and as computer programs. An example of a representation is Structured Modeling (SM) that offers different forms of viewing models that are customized as per user’s interest and technical capability. Another advantage of SM is that it supports and facilitates model integration.
The other dimension of IME, besides the schemata, is the library of model solvers (i.e. software, algorithms, etc.), which manipulates the model schema and solves particular model instances. IME is thus, also used as a tool to convert data from the model schema data structure to the particular format desired by solvers (i.e., SAS, Simscript, GAMS, Spreadsheet, etc.). Such "grand scale" IME models however, are not in existence today and are only a hope for the future.
Moreover, IME can benefit an organization by facilitating communication among models and thus, extend the utility of any particular model beyond the function for which it was developed. IME can further be used to make models consistent with organizational goals, as it is able to capture both financial and operational impact of specific decisions. This would enable an organization to close the gap between performance measurement at the time of investment and at the time of execution.
Organizations that intend to reap the benefits of IME, however, must recognize models as a valuable resource of the organization in the realization of its strategic objectives.
What To Do - Models Not Yet Designed
Following are the recommendations for models not yet designed:
Use cross-functional design teams: Include representatives from providers to, and customers of, the function and choose team-members with great care. Also, it is important to ensure sufficient availability of resources and the creation of an environment that is supportive of the cross-functional team;
Make the user the leader of the design-team: Although technicians are organizationally "above" the users, the technicians should be assigned to work and take direction from the user;
Develop the model within an IME: Use model schema representation like Structured Modeling (SM). The achievement of this third objective, however, is contingent upon successful implementation of the first two recommendations.
What To Do - Models In Process
Following are the recommendations for models currently under development:
Reexamine the model from a cross-functional perspective: If the development team is not cross functional, this mix should be altered. Also, attention must be given to providing sufficient resources and a supportive environment to the cross-functional team;
Convert the model to the IME: Develop appropriate schemata. To do this however, "reverse engineering" may be necessary.
What To Do-Models In Place
For existing models, "reverse engineering" is necessary to convert them to the IME. This requires developing the appropriate schema. As this is usually a time-consuming and costly process, an organization may want to choose only the most critical models to be reverse engineered and leave the rest untouched. Thus, a cost-benefit analysis may be needed to identify which models are most critical. However, in conducting such analyses, both quantifiable costs and benefits and qualitative impacts must be considered. By performing such an analysis, an organization is able to promote a cross-functional awareness and focus. Thus, in this process, the organization is able to evolve incrementally towards an integrated modeling environment.
In order to achieve a cross-functional focus a change is necessary not only on the "shop floor but also in the models used to support decision making, and by implication, changes in information resource policies" (p. 211). Further, successful implementation of model integration requires a high level of coordination and cooperation than before. Moreover, "model integration is not an answer to modern management problems, it is simply one of many tools that will help today’s managers make more informed decisions and maintain a competitive market position" (p. 211).
Elliott, R. K. 1992. The third wave breaks on the shores of accounting. Accounting Horizons 6 (June): 61-85. (Summary).
Hammer, M. 1990. Reengineering work: Don't automate, obliterate. Harvard Business Review (July-August): 104-112. (Summary).
Johnson, H. T. and A. Broms. 2000. Profit Beyond Measure: Extraordinary Results through Attention to Work and People. The Free Press. (Summary).
Martin, J. R. Not dated. Responsibility accounting compared to other concepts: Summary exhibits. Management And Accounting Web. https://maaw.info/ResponACCSum.htm
Martin, J. R. Not dated. What is responsibility accounting? Management And Accounting Web. https://maaw.info/ResponsibilityAccountingConcept.htm
McNair, C. J. 1990. Interdependence and control: Traditional vs. activity-based responsibility accounting. Journal of Cost Management (Summer): 15-23. (Summary).
McNair, C. J. and L. P. Carr. 1994. Responsibility redefined. Advances in Management Accounting (3): 85-117. (Summary).
Mintzberg, H. and L. Van der Heyden. 1999. Organigraphs: Drawing how companies really work. Harvard Business Review (September-October): 87-94. (Summary).
Parker, L. D. 1984. Control in organizational life: The contribution of Mary Parker Follett. The Academy of Management Review 9(4): 736-745. (Note).
Tiessen, P. and J. H. Waterhouse. 1983. Towards a descriptive theory of management accounting. Accounting, Organizations and Society 8(2-3): 251-267. (Summary).