This is the third article about the profit maximization feature in network design tools. In the first (and video interview – at the 3:55 mark), I discussed two big reasons why you shouldn’t use the maximize profit feature. The second talked about using the feature to debug models or for creative work-arounds.
Coming out of the first two articles, you might get the impression that I’m suggesting the business objective should only be to minimize cost.
Watson Says... |
 |
| Network design is a great tool to help figure out how to capture more market share and more revenue. |
 |
What do you say? |
 |
| Click here to send us your comments |
|
|
|
In fact, Terry Harris of Chicago Consulting sent us a comment that companies use network design to maximize revenues all the time. Terry cited Amazon’s ruthless focus on lead-time reduction to capture market share. Amazon is likely using network design tools to determine where to place their facilities so they can give you efficient 2-day service and even 1-hour delivery times.
I agree with Terry. Network design is a great tool to help figure out how to capture more market share and more revenue. I just don’t think the profit maximization feature helps you achieve this objective. You can do great work to design a network that maximizes profit or revenue while using the cost minimization feature. Here’s why I think this:
When it comes to lead-time reduction (like what Amazon is doing), your network design model uses a combination of cost minimization with maximizing the demand within a certain distance (or time) of a warehouse. OK, I cheated a little bit and threw in service level optimization as a feature—but this feature compliments cost minimization. This is actually a great time to use multi-objective optimization—see the first part of this article for exactly this trade-off.
But, even in these models, the demand is an input. Note that demand being an input doesn’t mean we aren’t trying to increase demand. But, for any given scenario you are going to specify what the demand is. You may enter your historical or forecasted demand. Or, you may enter demand based on demographics—basically a forecast of how well you could do if you could reduce your lead-time. Even though demand is an input, when analyze the outputs you are looking to pick a solution that you know will give you shorter lead-times (and improve revenue) with minimal incremental costs.
Even without lead-time, if you do the analysis creatively, you can use cost-minimization model to help the business maximize profit. You do this with careful analysis of the landed cost report (showing which customers and products are relatively more expensive) and by careful analysis of the capacity investments (how much do you have to invest to meet growing demand).
We worked with a fast growing company with two plants, one on the east coast and one in the southeast. They were growing fast on the east coast, fast in the southeast, and fast in the new Midwest and Texas markets. When they did their analysis, they realized that the needed to add plants or lines to existing plants to serve the Chicago area and the Texas market. By analyzing the investment decisions and using the model to drive the right conversations in the business, they decided to sell the Midwest and Texas business to another company and focus on their core markets.
Final Thoughts
Keep in mind that your model is just meant to help you make better decisions. Don’t get caught up in the features of the model. More times than not, the profit maximization feature will add more complexity than needed and may not answer the question as well as a cost minimization model- even if the business objective is to maximize profits or revenue.
Any reaction to this Expert Insight column? Send below.
Your Comments/Feedback
|
