Simplifying Landed Cost Modeling Series Part 2: Complexities and Validation

Simplifying Landed Cost Modeling Series Part 2: Complexities and Validation

Validating our cost models is essential. Although it can be tedious and frustrating, validation ensures a more accurate Cost Model Imageand useful product. We need to validate the relationships within the model function as expected; that the data inputs are correct, and that any assumptions are valid. Time and money can be wasted in cost modeling due to complexities. These complexities can be caused by:

  • Size of the model
  • Complexity of the model
  • Disconnects between the parties involved (creators and end users)

Below are five ways to overcome these complexities, smoothing out the validation process.

1. Take time to understand the data from those who create it. Model validation is meticulous and time consuming, and may involve people who are not connected to the project, as well as undefined processes. Try to connect with the individuals who are actually gathering data from the systems and processes. A 30-minute conversation can make the difference between obtaining the data you need verses reworking data collection (or worse - using bad data without realizing it).

2. Work closely with end users and have them validate results during model creation. Weekly updates and PDCA’s are essential to minimize the pain in validation. Your end users are typically closer to the work being modeled and can give immediate “gut check” reactions to model results (even during intermediate model creation). This collaboration can save time and effort by catching incorrect sub-results before they get buried into the layers of complexity of these finished models.

3. Simplify the model as much as the model purpose allows. Simplifying the creation of the model can help in the validation process (see Part 1 of this series) as manually working through the equations and data checking becomes easier.

4. Focus on the most important “drivers.” Depending on the usage of the model, you can simplify validation by focusing on the most important “drivers,” or data and equations that have the largest impact on results. A quick sensitivity analysis will typically find these drivers. Think of the ‘Pareto principle’ (80-20 rule). I commonly find this principle illustrated during projects. If validating the top 30-40% of the data inputs and equations gives you 95% confidence in the model’s reaction, then you may be able to scale back the time and effort of model validation.

5. Document your logic, equations, and assumptions. Good documentation can take precious design time, but becomes a gold mine for validation (and user training). In-line code documentation, pseudo code work-ups, equation explanations/development, and assumptions lists are just a few examples of documentation that can greatly aid in understanding a cost model during validation.

Written by Chris McLaughlin, Lean Deployment Specialist at LeanCor

Enhanced by Zemanta

Posted by LeanCor Supply Chain Group

blog author

LeanCor Supply Chain Group is a trusted supply chain partner that specializes in lean principles to deliver operational improvement. LeanCor’s three integrated divisions – LeanCor Training and Education, LeanCor Consulting, and LeanCor Logistics – help organizations eliminate waste, drive down costs, and build a culture of continuous improvement.

LinkedIn Google+



NEW: Resource Center

Lean Logistics and Transportation Management

From carrier management to network design, browse our best content – webinars, articles, videos, podcasts, and more. 

Visit the Transportation Resource Center

Subscribe Now!

Recent Posts

Supply Chain Brief