SAP Business Warehouse Simulation Introduction

Simulation plays very important role in SAP Business Process and helpful in forecasting the business future requirements.Simulation is a procedure to analyze dynamic systems. In Simulation experiments are carried out based on a model of the reality, in order to gain insight regarding the real situation.Business simulations allow you to evaluate your business, identify key issues driving your processes and capture your company's specific business challenges Simulation models may range from simple linear equations with one unknown factor to sophisticated models that incorporate all known influences. Usually the focus of a simulation lies on aspects, which have the most impact on the simulated system, whereas minor important factors are left off or simplified.

There are different methods that support simulations:

Forecasts: Forecasts examine trends in the presence and past and prolong those into the future. There is no guarantee that those trends will last, however besides day to day turbulence's there are a lot of development processes that are more linear than they look at first sight.

Driver-based simulation
Model-based simulation
System dynamics: Small events can cause unforeseeable processes. Watch the stock markets (or the weather) and you can see the effects any day. System dynamics is the attempt to predict future situations based on all available information and specific theories.

Driver-based Planning

Planning is often executed on a very abstract detailed level (e.g. cost elements). When a cost center manager is asked to submit the expected costs for salary, travel, office equipment, overhead costs, etc. that information is usually based upon an information behind those figures (e.g. based on the number of employees working in his department). The challenge is to translate data from across business functions into critical information that tells a company what drives their business plan Driver-based planning takes those dependencies into consideration. It simplifies planning on an aggregated level and enables simulations based on different assumptions about the drivers. Reactions on changing business conditions can faster be decided upon by using changed driver values.

Methods to support driver-based planning are:

Attach documents: people are basically asked to submit a supporting narrative outlining the assumptions that are behind their numbers.
FOX Formulas: model inter dependencies of drivers and related plan figures
Planning sequences: Re-plan based on a predefined sequence of planning functions
Versions: Simulate different scenarios without overwriting
Reference Data: use relations and inter dependencies from the past

Model-Based Planning

The planning process spans the whole organization. Decisions in one area of the company influence processes in other areas. These relationships may be executed sequential in the initial plan. For plan revisions the dependencies of the different plans may be modeled according to specific rules. That way it is possible to only change some plan assumptions and recalculate the full model.

Example:

Sales planning and production planning must be aligned so that based on the sales volume and assumed time for the demand the goods must be provided by production at the right time. Sales and production need resources in order to achieve their targets, e.g. sales representatives or production employees which influences headcount planning.

In addition to the integration aspect every department must conduct its internal planning process that in itself also may be model-based (e.g. calculation of capacity restrictions or scheduling in production planning)

Model-based planning reconstructs those inter dependencies and allows to simulate different scenarios based on alternative plan assumptions.

Methods to support model-based planning are: .

FOX Formulas: map data models (e.g. master data) and transfer data
Forecast Functions: predict future trends for value drivers which influence your overall model (e.g. sales growth)
Multi-Planning Area: Exchange data between different plan cubes
Planning sequences: Re-plan based on a predefined sequence of planning functions
Versions: Simulate different scenarios without overwriting
Reference Data
Aggregation: Every planning process involves a lot of detail which cannot be mapped to another process. However there are similarities (e.g. Sales volume = Demand volume, time period) that may be used as an integration level. The aggregation in a planning level allows to exchange the integration
values.

Limitations of BW-BPS Regarding Simulation

Circular references (A+B=C, C+D= A)

Excel formulas: no circular references possible
FOX formulas: circular references possible through planning sequences

Iterations

BPS offers no predefined functionality for iterations
FOX formulas: iteration only by re-execution
Planning function type exit: iterations possible

Graphical display of relationships

Dynamic Simulation

System Dynamics (SD) is an experimental approach to System Analysis. It is a way of understanding complex systems and modifying or changing them in some way. It also an approach for validating and assessing the consequences of implementing analytical (prescriptive) models or recommendations of a case study report.

System Dynamics is both

a theory of structure in systems;
an approach to policy design.
System Dynamics is comprised of two concepts:
Feedback Theory that provides general guidelines for organizing system structure.
Computer Simulation that provides a means to deduce the behavior arising from a particular system structure.
System dynamics is concerned with the construction of graphical and mathematical computer-based models, with detailed descriptions, that tells us how the conditions at one point in time lead to subsequent conditions at later points in time. The constructed model can then be simulated and its behavior observed over time.

To sum it up,

System Dynamics is about studying complex and dynamic systems - systems which change over time.
System Dynamics is about finding the 'why' (cause[s]) and 'how' (pattern) of system changes.
The main benefit of system dynamics is to analyze single complex