My first introduction to modeling was way back in the Eighties.

 

My job was to recover £15 million (some of which had been owed for over 30 years) from an overseas Government and figure out a way to prevent such a situation reoccurring. Over a period of months we managed to recover the debt but designing and getting agreement on a system for the future was even more of a challenge. Fortunately, I had a friend who worked for a commercial shipping line. He explained how fundamental Letters of Credit were to the import/export industry and the light bulb went on.

All I had to do was explain the concept to the “powers that be”, find a shipping agent and a bank willing to work with all parties to get the ball rolling.

Nobody at all in my organization knew anything about Letters of Credit so I had to find a way to communicate how the process would work in order to convince the decision makers to support the project. At that time I had no training whatsoever in process analysis and redesign and no tools at all but I figured that a picture would be a good way to convey what I was on about.

I got a pencil and the biggest piece of paper I could find and started to draw the procedure based on discussions with a cooperative shipping agent, a couple of international banks and the myriad of departments from both Governments. I soon realized that I would need a bigger piece of paper (and an eraser), and (more importantly) a system.

A colleague suggested that I use a plastic stencil consisting of a variety of shapes (rectangles, diamonds, circles, etc.) at least to neaten things up a bit. The stencil worked wonders and represented my first introduction to a modeling language, in this case IDEF0 (Integrated Computer-Aided Manufacturing Definition), and to business process improvement.

In those days, we did not even have PCs, so my method was to use the stencil to draw the required shapes, write the process task descriptions by hand, stick the shapes on paper (I had now progressed to a roll of paper), connect the shapes together by lines drawn with a long ruler, and pass each iteration to the typing pool to figure out how to produce acceptable output.

Needless to say the entire project took months to complete and cost me many boxes of chocolates for the typists (who began to dread the sight of me). However, it was a success, as eventually all parties signed up for the new and improved way of doing business and goods and money flowed as per the process design.

Apart from giving away my age, this story illustrates that organizations have been using modeling as a means of improving the way they work internally and interact with other entities for some considerable time. The concept remains the same – just the tools, skill levels and general awareness of the techniques has changed.

Since my little story the world is has embraced technology and in spite of the rise (and fall?) of business process reengineering, the Quality Movement continues to march forward relentlessly with a plethora of standards and methodologies most notably ISO; Total Quality Management (TQM); Statistical Process Control (SPC); Six Sigma; Lean and a range of process mapping techniques including: Event-driven Process Chains (EPC); Entity Relationship Diagrams (ERD); Value Stream Mapping (VSM), and so on.

 

All of these business process improvement techniques have their place and have been used with varying degrees of success in the pursuit of organizational and operational excellence. However, many rely on process maps as the basis for capturing flow and improving processes. Having personally used just about every process mapping technique under the Sun, I contend that in some ways they have not progressed very far from my rudimentary paper and pencil approach. In my view, none of the techniques mentioned can come close to the power and effectiveness of process modeling and simulation.

Now it is time to declare a vested interest. Steelnet Solutions, is an authorized reseller of ProcessModel™ process modeling and simulation software and is the only certified trainer for this software in Canada .

Consider the following:

  ·   Process models move; With simulation software you can see what is going on. You can see an object (document, package, vehicle, person, etc.) moving through the process. The mere fact that a process model is animated is infinitely more powerful than a static illustration. People can finally understand the impact of their actions on others. You can’t do that with a process map.

  ·   Process models can analyze the process for you; There are many different simulation packages available but the better among them can automatically generate and display in reports and graphs at least a standard set of critical data (typically by average, min-max and standard deviation) e.g.:

 

- Process cycle time

- Activity time

- Process/activity cost

- Time in queue

- Resource utilization

- Value added time/Non value added time

In addition, simulation software will usually allow the user to define their own data requirements by assigning attributes and variables within the model. These are used to describe or track activities and states in the system. What this means is you can build a model that reflects your reality and analyze the process in a way that suits your unique requirements. I do not know of any mapping technique that can do this.

The only way I have been able to even get close to analytical capabilities of simulation software is by capturing the required data, entering it into Microsoft® Excel or Microsoft® Access and analyzing it discretely. Although inherently inefficient, this procedure actually works but the fact that the data lives apart from the process map becomes a huge problem the minute a change is made to map. Conversely, any change made in a process model is instantly reflected in the statistics generated once the model is simulated because the model and the data reside on a single database.

 

 

  ·   Process modeling and simulation makes “What if” analysis and process optimization easy. One of the most basic requirements of any process improvement exercise is to consider the impact of “change”. As previously stated, process maps render even simple experimentation an onerous and time consuming task. With simulation software a user can change any value or combination of values and quickly see the results. In most of the better packages you can design multiple experiments and automatically compare the results. In fact, this scenario based “What if” analysis can be taken to another level of sophistication if the process modeling and simulation software incorporates an optimization engine (such as SimRunner) which automatically seeks an optimal solution.

  ·   The ability of modeling and simulation software to handle the inherent complexity associated with many processes leaves traditional process mapping for dead. Here are some examples:

- Conditions. Here the flow of process changes based on a defined condition e.g. time, priority, resource availability, probability, etc.

- Interrelationships and dependencies. Processes rarely live in isolation and although it is possible to illustrate relationships and dependencies using process mapping techniques it is a lot easier more effective to do so using modeling and simulation software. For instance, how can a process map communicate and calculate the cost of delay?

- Variability and randomness. Real-life processes may follow a defined path but their behaviour varies depending upon events. Simulation software is adept at handling variability e.g. variable ordering patterns depending on the day of the week or the time of day. Also, in the real world, events can (and do) happen randomly. Such randomness can be modeled easily using simulation software.

- Concurrency. Organizations do multiple things simultaneously and so depicting this in any detail is a challenge using process maps.

- Speed. Whether you are mapping or modeling, in my experience, most of the time and effort is expended in gathering reliable information about the process. However, once that task has been completed, the actual development of the requirement is much faster and easier to accomplish using the more modern modeling tools given their advanced and user-friendly interfaces.

The most common criticisms put forward in respect of process modeling and simulation software

are:

 

  ·   It’s too complex and hard to learn. In truth, anyone who has any familiarity with Windows modern simulation packages are surprisingly easy and fast to become learn. You do not need a brain the size of planet to get great results. You do not need to be a programmer.

 

  ·   It’s too costly. Over the years, with the emergence of p rocess modeling and simulation packages designed specifically for business users (rather than academics or statisticians) the solution has become more affordable and accessible. That said, the ease with which

dramatic process improvements can be designed using this software makes modeling and simulation a zero cost investment (provided that the results are actually implemented).

In conclusion, process modeling and simulation software is the most effective tool for improving business processes, communicating ideas, and solving complex business problems.

Here is a selection of resources for further consideration:

http://www.acslx.com/

http://www.arenasimulation.com/

 

http://www.createasoft.com/

 

http://www.flexsim.com/

 

http://www.processmodel.com/

 

http://www.quask.com/

 

http://www.simprocess.com/

 

http://www.simcadpro.com/

 

http://www.simul8.com/

 

http://www.vissim.com/