The purpose of a simulation is to find a simple way to manage your expectations of what a project will deliver. We know that there will be a lot of testing involved in that delivery process – however, fundamental issues cannot be fixed in testing that only can work effectively once that project reaches its final stages.

Sometimes we create a mockup application in a real-world setting to drive high-volume transactions through that application environment as a good way of confirming that this environment is capable of handling a given workload. The “TILT - Timed Input & Load Testing” course shows a method for creating such tests in a physical environment. SCIENCE can be used to simulate such tests in a logical environment that requires no more investment than some time and effort to build a model in Excel™.

Simulation is never the real thing – and you may make assumptions that might not be real-world accurate – but you owe it to your clients to eliminate the more obvious flaws in the proposed project objectives. We would not build a life-size aircraft without first simulating characteristics in a computer, then wind-tunnel, then scale model, with feedback to the designers until finally it looks like a safe bet to build a full-scale airplane. There is nothing stopping us from doing what with applications, until we see a prototype that proves the concept before we authorize a project to implement the proposed solution.

There are three aspects to the SCIENCE engine that can be combined for exceptional power and flexibility in how you want to evaluate a future project. The following areas of capability will be reviewed:

Combinations driver

The fundamental logic is based on injecting sample values into some calculation model, similar to our financial simulation tools for financial analysis in business cases. This is probably the most common use of the tool for problems outside the domain of financial analyses.

Transactions driver

In order to perform capacity simulations we need to drive workload that reflects what the simulation needs to process. This is similar to what we incorporate in “TILT - Timed Input & Load Testing” except that we do not generate workload for external consumption. Instead, these timed transactions will be processed through the SCIENCE engine using a model of the environment so we can keep a close look at performance details (based on the operating assumptions we want to test for).

Simulation Functions

The logic is enhanced with built-in functions that can be triggered to decide what the next step is in the simulation, such as finding a next node in a processing network based on load sharing (like finding what looks to be the shortest cashier line in a supermarket). By implementing these kinds of behaviours we can see the results in a queuing model.

Basic (Linear) models tend to focus on finding the optimum balance of input parameters to figure out what values to use in order to maximize production profitability, for example. In this course we will go through a basic model as an exercise, and/or use a challenge that the student has for which we can construct a basic model. The purpose of this exercise is to introduce the analytical steps involved in establishing how to create a model analogy for a real-world challenge.

The key to this part of the course is lateral thinking, imagining how a particular physical situation can be reduced to its core elements that fit a logical model for which we can establish an Excel™ worksheet with all the elements and capabilities to calculate a single performance instance. It is important to determine the accuracy of these calculations to make sure that we do not simply automate “garbage-in/garbage-out” at a rate that makes you dazzled, we need the generated results to be reliable. If you don’t have precise data you can still simulate so long as you make it a reference that others can see (such as the cost of equipment that may change at some point in time, that you do not want buried in a formula).

Based on the numerical results from a model we may be able to draw certain conclusions that can impact on how we approach the challenge of the project to be implemented. The key is that no animals are hurt in testing various hypotheses no matter how many cases and/or how long we drive the analysis until we are confident that the project is feasible. Usually we know what we want to prove or disprove – the purpose of a model is to confirm the hypothesis rather than to discover a whole new knowledge area because you program the model accordingly. We make sure that the fundamentals of modeling are well understood as a result of running through a set of examples.

Complex (Queuing) models tend to focus on finding capacity issues or limits in a proposed production infrastructure, for example. In this part we review a number of built-in SCIENCE functions that assist with a number of typical simulation functions, like queue management, to give the system its enhanced powers of pretending to be a physical process of some sort. In this case, what we look at is the performance of queues over time – instead of sorting the output columns the SCIENCE engine will take a snapshot as stated intervals (seconds since the last snapshot) until the maximum number of columns is reached (or until all transaction data have been exhausted).

We read the output from this type of simulation differently as we see how queues build, or if the model “stabilizes” at a given backlog, or what the impact is of having multiple parallel queues (servers, etc.) to balance the workload in a given system. We can use colours to highlight when a given minimum workload is reached, when it reaches 75% of capacity, and when it reaches 90% or more. When the workload reaches a limit that can be defined in the user-supplied parameters that can be shown as a bottleneck in the proposed system. Other kinds of measurements or observations can be accumulated to learn from the inputs before any decisions are made on how to proceed with the project in question.

Capacity analysis or throughput analysis may present a challenge in how we construct a reasonable proxy for the real-world process that we want to simulate. This is especially true if we do not have a static case to compare against (for example, a time and motion study that establishes the performance under a static condition, or a formula that shows what the performance is under a curve that crosses several static observations so we can adjust the model dynamically). These concerns will be part of the course discussion, but it is difficult to anticipate all possibilities.

SCIENCE allows you to save output data in a results column for each individual case, and it will keep the columns ordered based on your sort criteria, so that you can see what the best alternatives are. Results must never be a single set of parameters – you need to know the freedom or sensitivity for individual parameters that you can then control in order to optimize a production environment. SCIENCE can transfer specific data to a secondary worksheet, and produce graphs to reflect what the result of a given analysis was – this will help you to present a business case for (or against) a potential project initiative.

Because the data are always maintained in an Excel worksheet there is infinite flexibility in how you will use the results in various reports that a client may ask for. Always remember that the client does not have the in-depth understanding of what your analysis covered, so that you must enhance the numerical results with a clear explanation of what it means. Typically you will write your findings in a report and you support that with the analysis results in tables or in a graph (here is where you can use the standard Excel™ functionality to your advantage). You may have to repeat the simulation several times to change the hypotheses used when you are not sure of key variables, to determine the risk of implementing a solution when key information remains outstanding.

Understand the limitations of simulation: you can typically avoid the risk of a total disaster, but it can be difficult to know exactly how things will turn out under real-world conditions. Also, the kind of simulations we perform are only one aspect of what may be available to help you, such as BIM (Building Information Modeling) that lets you step through a building construction to make sure that what you want to do is actually possible, or working with models of boats or airplanes to see how well they actually stand up to real-world conditions. Understanding concepts of simulation will help you to appreciate many other tools that will help you to reduce the risks inherent in your next project.

Learning Formats       PMAP_SCI

This course is currently available in a classroom setting (public or company private) with approximately 30 contact hours (5 days).

PDF – Certificate Of Completion

Each course offers a certificate of completion that identifies the course, the student, and a brief description of the course. To receive a certificate the student must have attended at least 80% of the course sessions. This personalized certificate is forwarded to the student by Email.

PDF – Course Notebook

Each course includes a notebook in PDF format that provides the minimum knowledge the student must master in order to obtain the certificate. In the notebook you will find references to other study materials. Students receive the notebook by Email when their registration is confirmed.

PDF – Program Overview

An overview of this study program can be downloaded from the website by right-clicking on the program link on the enquiry page.

PDF – Current Training Schedule

A list of upcoming training sessions can be downloaded from the website by right-clicking on the schedule link on the enquiry page.

Registration – Service Providers

To register for any training course please look on the enquiry link page of your service provider (from where you accessed this website). On the page you will find a registration request form where you can order the course that you are interested in. The availability dates will be provided to you, along with payment instructions if you decide to go ahead.{{ENDSKIP}}