|
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
     |
||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||
|
What's The Question? by Frances Collins Before the decision to use simulation is reached a question will have been identified and defined. The importance of this stage in a project often gets overlooked. In fact as the simulation modeller you might not even be part of this stage of the project. What's the Question? The question definition stage is the very first step in any project. I think it begins from the second someone perceives that a process should work in a better or different way. This could be because they feel it's inefficient, or perhaps cuts have to be made or because policies have changed. The goal of where they want to get to will more than likely be defined correctly but how they get there becomes the question. As the simulation modeller you'll often be delivered the question in terms of the solution. For example I worked on a project where we built a simulation of an application process that had a backlog of months. We were tasked to discover how many additional staff would be required to get rid of this backlog. Problem: backlog, question: how many more staff. But was that their real question? Once data collection began it emerged that they had a 75% rework rate. (This was new information to management!) Eliminate this and the majority of their backlog would have disappeared. Unfortunately my project question was not redefined, the project was completed and the number of additional staff needed found. Was this a good solution to the wrong question? I believe so. Personal Questions Why was the right question not identified in the beginning and when it emerged why was the question not restructured and the examination begun again? All questions are personal. A question will be perceived differently by different people. Both because they have different ways of understanding what is going on around them, and also possibly because they have different interests and responsibilities, which lead them to different concerns. For example lets say a manufacturing plant installs a new million dollar system whose throughput is supposed to be 500 widgets per day. But it never produces more than 250. The plant perceives the question to be with the design of the new system whereas the designer of the system sees it as an operational issue. The system is capable of handling 500 per day as defined in the specification but the plant produces 3 products of different size, each product type goes through a different sequence and has different processing times. The plant operator isn't scheduling orders but just placing them on the system as they arrive. The excessive changeovers and queues building up in front of the machines causes the reduced throughput. So is it a design flaw or an operating issue? Probably both but how the question is tackled will depend on who has ownership of the question and how they view the question. Also people don't like to be criticized, especially in the working environment. People naturally become defensive when it 's suggested that they may not do their job in the best way. They also often go on the offensive to find someone else to blame. Also depending on who's asking the questions they will probably give different explanations. Would you tell your boss you can't cope with your workload? This means that the information needed to establish the question can be very difficult to get at and that you often receive biased information. The inter relationships and power basis between the parties involved can also be a significant factor. Would you want to go to your superior and tell them there was a 75% rework rate in your department? Neither did our client! Setting Boundaries The second most dangerous area of question definition is setting the scope. It is very easy to over simplify or over complicate a question. If you over simplify the question you run the risk of missing out key issues and factors. This could mean you provide a solution that may be invalid in the real world or will solve one question but create others. If you over complicate the question the study can become too complicated. The additionally complexity could mean that the project will take much longer to complete and you may loose client buy-in, or miss deadlines. The complexity could also make the model difficult to analyze so that no clear solution can be found. The boundaries of the solution also need to be considered. For instance in the example above the 75% rework rate was largely caused by customer application forms being completed incorrectly. The processing department had little influence on this factor. The marketing department would need to tackle this issue with a better form design but they were not part of the project so this was not considered within a valid solution. It was also a solution that would take much longer to implement. Renegotiating the Question So what do you do if you haven't been involved in the question definition stage of the project and you feel it's been defined incorrectly. There are 3 options. Be coercive. Use some of your power and position as the expert to tell the client what question you think they should be investigating. Depending on how set the client is on their version you may be able to carry this off but if you do you'll need to be 100% certain that your approach is the right one! Be empathetic. You completely ignore your instincts and work only to the definition the client has set out. This is the easiest approach and means the client will feel they're getting exactly what they want, but what happens at the end of the project if the question is unresolved or has failed to produce any significant improvement. The client will more than likely blame you after all they hired you for your experience and to get expert advice. Negotiate. I think this is the best solution. Here you negotiate with the client to redefine the question. This is more likely to work as you can point the client in the right direction but the client also feels that you're working on a question that they feel interested in. If there are multiple clients with differing views you might try running a Workshop. (See our newsletter from October 2003 for more details on running workshops.) More Reading I have briefly dealt here with just a few of the issues that make problem structuring complicated. I think it's a fascinating area of discussion and definitely believe it's an art as well as a science. If you 'd like to read more on this topic here are some references you might find interesting. Eden C, Jones S and Sims D, Messing About in Problems, Pergammon
1983. Our website is always being updated with new support resources. Here are details on some of the downloads that were added this month. New Service Pack Version 10 Build 630 New Learning Zone Simulation - Changeover After Completing Work
Item x Delete Work Item in Work Center or Work Entry Point Don't Change Value On Reset
Upgrade For:
This feature is only available
with SIMUL8 Professional 
If you have any questions generated from, or comments about this
newsletter then email Newsletter@SIMUL8.com
|
|
||||||||||||||||||||||||||||||
