Litt's law-Little's Law

Little's law, derived from queuing theory, is the most widely known law in IT system performance modeling.

Litt's law revealsLead time(Lead time),Number of products in process(Work in Progress,WIP) AndThroughput(Throughput.

• Lead time-lead time: Only the time period between the request entry to the system and the completion of the Request acceptance. The lead time is measured based on the elapsed time (minutes, hours, etc. A request can be a requirement, a user story, an exception, material, or a request from a user.
• Number of products-Work in Progress(WIP): Number of requests being processed (unit of work). The number of requests being processed indicates that these requests have already entered the system but have not yet been output.
• Throughput-Throughput: Number of work units that leave the system within a certain period of time, for example, 3 user stories/day.

The conclusion of this law is quite interesting and important:

• The larger the WIP, the longer the lead time, that is, the longer it takes to complete the started work. In other words, to meet the development or service deadline, we must reduce the number of work-in-process or finish work-in-process before starting a new job.
  
  However, in many cases, the opposite is true: to make the entire project "run" faster, the team starts to process more tasks. Another reason to maintain a large amount of ongoing work is to achieve high resource utilization.
  
  For whatever reason, assuming that the throughput does not change, the increase in work-in-process will increase the time required to complete the work-in-process (that is, the lead time) at the same time.
  
  Even though it seems intuitive, it should be noted that lowering the WIP helps meet the SLA and reduces the development lead time.
  
  
• Focus on reducing lead timeHelps identify useless activities that have been carried out. Eliminating these useless activities has two positive effects:
  (1) Eliminate waste in the process
  (2) reduce the total wip, further shorten the lead time and bringMore efficient development.
  
  
• Higher throughput, shorter lead time
  
  There are many different ways to improve performance: self-built value-added activities (self-built non-value-added activities are equivalent to self-built waste), improving processes and adding more resources. If you decide to add more resources, you need to observe the overall lead time, because adding additional resources will increase the number of products at the same time.
  
  
• Each Lean initiative attempts to minimize waste and shorten production cycles. Shortening the production cycle is equivalent to shortening the lead time. Minimizing waste includes analyzing the current inventory and reducing the inventory, which is equivalent to reducing the WIP.

Why is this law important to project managers?

1. Litt law is a tool that understands the real performance of a software development team or software operation team.
2. Provide predictability
   For example, if we have to implement 50 requirements and the average team capability is 5 requirements/week, we need the following time:
   50 requirements/5 requirements every week = 10 weeks.
3. It is revealed that the larger the number of jobs, the longer the processing time (lead time.
4. Explains why multiple tasks lead to delays rather than acceleration.
   It is often believed that parallel execution of multiple tasks can improve productivity. Therefore, a common practice in many companies is to assign multiple tasks to one person. However, unlike machines, humans are not good at executing in parallel. Increasing the number of work processes increases the modification time and rework time of a task to reduce the throughput. In the end, the execution of the work is not enough time, and the work that has started but has not been completed begins to pile up.
   In short, Litt's law is conducive to finding a balance between the work in progress and the lead time.
5. This provides a foundation for achieving the optimal WIP limit.If the WIP limit is lower than the optimal level, there will be underutilized resources and low performance. If the WIP limit exceeds the optimal level, the unit of work begins to accumulate in the queue and the performance is also reduced.
6. Helps you understand the impact of blocking a job or resolving errors on the project or service end time. In both cases, the throughput is reduced and the lead time is increased.

Important conditions for applying Litt's Law:
Litt's law is very useful, but in addition to understanding formulas, you must understand the conditions that must be met when using Litt's Law:

• Average value is used for all parameters: average lead time, average wip, and average throughput.
• The Unit must be consistent: that is, if we measure the throughput by one week, the lead time must also be by week and WIP.
• The system must be stable: that is, all the work that enters the system will definitely leave the system; the total WIP will be retained during this period, and the average work arrival rate is equal to the average work exit rate.

In short, correct use of Litt's law can help to achieve stable workflows and improve project and IT service predictability. WIP is a key factor that affects project performance and the time required for software development or service completion. Limits WIP to reduce the lead time and reduces the waste of work flow.

Litt's law-Little's Law