Preliminary Question before a simulation Project
To begin with a simulation project, there are a few things that we need to capture to ensure the success of the project.
This will set the boundary, the purpose and the approach of the project.
- What is the question that we would like to answer in this project?
e.g. We are unsure if we are able to meet the demand when it picks up by 30%?
- Are there a specific problems that you want to solve?
e.g. We do not know the reason for not meeting desired throughput？
- We need something to show and tell our customers what we have?
2. Problem description:
An accurate description of the process will help one to determine what is needed to be included and excluded in the model. We need as much detail information as possible to fulfill the objective of the model. Too much information will overload the model, slow down the run time and complicates the matter.
i. Describe your process flow or materiel flow
- A flow chart along with a step to step description is very important for one to understand the operation. Information needed will be
a. What kind of process? e.g. Loading/unloading/transporting/processing/storage/decision making?
b. How it is being done? e.g Piece by piece/batch?
c. Who is doing it? People/equipment?
d. What is quantity of resource (out of the pool of resource) need to do it? e.g. two operators out of 10 operators/ one Sr Nurse (out of 2 that has the same skill set) and one respirator (out of the pool of 5)
e. Priority rules (if any)
ii. Starting and ending point:
- The starting and ending point will define the scope of the model. For starting, question such as "Where do we begin?" If the answer is "After the outgoing pallets being sorted.", then the model will only after the sorting process. For ending, question such as "Are we interested in the truck movement after it is being loaded?". If the answer is NO, then the model will end at the dock. Once the truck is loaded, it can be disappeared from the model.
iii. Arrival into the model:
- We need to describe how the material appear inside the model. Question such as"Is patient arrives according to appointment?" If the answer is YES, then the patient will arrive at the registration according to certain preset time schedule.
iv. Cycle time:
The time that material is delayed at a certain process. A certain assumption can be made if the information is not available at the beginning of the project.
v. Floor plan: The floor plan helps one to place the resource at the right location in the facility. The floor plan should be in dwg or dxf format. The drawing unit should be sync with the modeling length unit.
vi. Decision variables:
- The information to be manipulated during the "what-if" analysis. It could be the number of operators, down time, lot size and others. Sometimes a setting up a table to store such information will be easier for handling the information.
In order to answer our question, what kind of information do we need to be extracted from the run of the model? E.g. To answer if the production system is able to meet the demand, we will need to know what is the throughput from the model. Sometimes, there is no direct match of the information such as where is the bottleneck? We will need to extract information such as the Avgstaytime (average time that the patient or material stayed at that queue) for the buffers and making an inferred conclusion after the comparison.
In order to evaluate the performance of the model if it is able to fulfill the requirement, a certain set of evaluation criterion is needed. For example, do we have enough operators? An assumption of 15% of average idleness in operator during the whole simulation is made. So if all of operators has less then 15%, it will indicate that the operators is running tight. More operator is needed.
How successful is the simulation project will depend on clear is the scope and objectives are being defined? No point spending more time to model individual part movement if the stakeholder only want to know the overall operation. Having said that, there are also stakeholders wants everything in the world, then the next question is how much time do we have? If you have so much time till the end time, then you can "play" around with the model until the stakeholder is happy. However this is normally not the case. Most of the time, they will want the result "yesterday". Then you will need to discuss with them what is to be considered and what is to be ignored in order to meet the project schedule.
Discussion.... discussion and discussion is the WORD!