Monte Carlo Simulation - How Many?

Content:
1. Introduction to "Monte Carlo - How Мany"
2. How to set up the "Monte Carlo - How many"?
3. How does this chart work?
4. What's that telling us?
5. Controls for this chart - customize your view

You can find more information about the Analytics module in the dedicated article.

1. Introduction to "Monte Carlo - How Мany"

The "How Many" Monte Carlo simulation aims to tell you the minimum amount of work items that you can get done within the next “x” days. For example, you can use it if you’ve got a deadline scheduled for July 7th and you wish to know how many tasks can be finished by then.
The simulation will run a lot of trials based on your historical data/throughput. It can run up to 1 million trials to create a distribution. The chart will be able to tell you with a probability of 50%, 70% 85% and 90% how many work items, you will be able to finish by a certain date.

2. How to set up the "Monte Carlo - How many"?

From the left side of the panel, you can control which data you can see on the chart within a predetermined time frame. 

First, you select the workflows that you want to analyze. The simulation takes into consideration all cards from the selected workflows. Note that you can add single or multiple workflows.

Second, you need to select at least one of the following fields: "Created At", "Start Date" or "End Date" and configure their value.

Note: you can't run the analysis without selecting at least one of the 3 fields. You can also select more than one of them for a more precise configuration of the dataset.

In addition, you can apply extra filters (Add cards filter). For example, you may want to see only cards with high priority or cards that are of a certain type. 

The Monte Carlo simulation consists of three charts:   

• Throughput basis (1) 
• Throughput Navigation (2) 
• Monte Carlo (3) 

The horizontal axis of the "Throughput Basis" (1) is a representation of time, while the vertical one shows the daily throughput. The high points of the charts represent the maximum throughput of any of the days in the chart, while the low points represent the minimum number of tasks that were completed. Hovering over the dots will give you summarized information about the number of tasks that have been completed that day. The line between the dots visualizes the rises and drops of your team productivity. 

The Throughput Navigator (2) allows you to zoom in and out of a specific interval within the selected time frame. This way you can generate different Monte Carlo simulation without having to reset the general time frame every time. 

Monte Carlo (3), which is the bottom chart is direct visualization of the results of the simulation in the form of a probabilistic distribution. The horizontal axis visualizes the total work items that will probably be completed within the selected future date. The vertical axis shows how many instances of a certain result occurred during the trials.

3. How does this chart work?

 (1) Imagine you have a deadline on July 7th. You have to select the date from the respective "End date" calendar picker. 

(2) Activate the percentiles (Select all) to get the probabilistic view of the third chart. 

(3) Run the Trials (up to 1 million) 

4. What's that telling us?

The third chart will be able to tell you with a probability of 50%, 70% 85% and 90%, how much work items, you will be able to finish by the deadline. The simulation shows you that there is 85% probability that by July 7th, you will finish 121 items. If you want to be more certain go to the 95% percentile, where the number of items is 103. 

Practical Tip: 
You may doubt the results if your work items are different in size or some of them are too complex.
That's why it is so important to keep your cycle time smaller. You have to do right-sizing and try finishing work within a certain period of time. If a task is too big, break it down into smaller pieces of work and flow it through the system. If you do this consistently, the forecast will be more accurate and your process much more predictable. 

Important: 
The analytics use one calendar day as the lowest possible time unit. 
For example:  The card was created 30 minutes ago and was moved to "In progress". The system shows 1 day for this card.  The card was created late yesterday at 23:00 P.M. and you take a look at the chart in the morning at 9:00 A.M. (i.e. around 10 hours after the creation). The system will show 2 days as the card has been created yesterday and is still being worked on Today. 
The Calendar time unit is used for analytics due to the fact that most of the modern Kanban systems are driven by client requests. As such, clients expect exact delivery dates or ranges that are calendar-based. They do not care about holidays, working hours, man-days, cycle, and lead times. 

That being said, the How Many simulation chart allows for simulations to be ran for working days only - this option can be enabled via the Simulation Control options.

Let's assume we have cards data for a full week (Monday-Sunday) and the data is as follows:

If the work days option is not selected, the simulation would be executed for the total number of completed cards for the full week i.e. 3, 5, 2, 1, 3, 1.

If the work days option is selected (e.g. Monday-Friday), the simulation would be executed for the number of cards corresponding to the days on which they were completed i.e. 3, 5, 2, 1, 3 and the one card completed on Sunday would not be taken into account.

Note: if the working days (Mon-Fri) are selected, the weekends would also be disregarded from the Begin-End date selections.

For example, if you set the Begin date to 01.01.2021 and the End date to 01.02.2021, the simulation would be executed based on items completed during the 21 working days of January 2021.

There would also be a 'Disqualified Items' section in the graph showing a list of work items that have been excluded from the simulations - in the above example, that would be the card which has been completed on Sunday.

In case that you need precise cycle and log times for internal reporting, use the Advanced search capabilities with the different charts views and widgets.

5. Controls for this chart - customize your view

You can use the controls to change the Monte Carlo How Many simulation view and to apply additional filters. Please, check the short video below. 

• Throughput Chart type - some users prefer Bar chart to visualize the throughput instead of a Line chart. 
• Percentiles - select and deselect percentiles to get the probabilistic view. 
• Layout - you can remove some of the charts and make only the simulation visible. Moreover, you can add a calendar that visualizes the begin and end date for your prediction. 
• Item Filter - apply additional filters to run the simulation for tasks that have a specific location or tasks that gain certain property.  Choose values to select which items are available to this chart.