Late last year, I was notified by my mechanic that my 1999 Isuzu Rodeo was going to die soon. Isuzu doesn’t make commercial cars anymore, so the costs required to find the non-existent parts for the repair were going to rise into the $1000s… which worried me, since my car wasn’t worth $100…
So, what do you think I did?
The logical next step…I found an affordable replacement over the next few months, but there was one significant problem that kept showing up when I was looking for my next car; I didn’t know anything about cars!!! How was I supposed to know if I was getting a good deal?
Thankfully, I had some time available in my work schedule and had some related projects that allowed me to indulge my curiosity. I started creating a template for commercial automotive vehicles, and I like the results so far.
Before I explain what we did, let’s make one thing clear. What I’m about to discuss is NOT how the commercial automotive industry usually addresses production costs for their vehicles. According to my coworkers with decades of experience in the automotive industry, what manufacturers do is estimate the expected sales volume of a certain vehicle, determine the selling price they feel the market accepts for that vehicle, and then design the vehicle to not exceed the estimated cost. Basically, the process goes in the opposite direction of how parametric estimation projects are done. As a result, the purpose of this project was to research and experiment in order to answer three questions:
During the cost research team’s investigation into how to model various cargo and logistics vehicles, we found an academic report that had detailed weight breakdowns for different car types at the subsystem level. These were based on various dismantling reports and were amalgamated to represent what you would see in a typical sedan, SUV, or pickup truck respectively. The weight breakdowns assisted in creating the PBS of the template to address major subsystems, and the complexity values (cost/lb) were based on information PRICE has on related projects. The first draft appears as the following:
I then looked through financial statements for the major car manufacturers that have plants in the United States such as Ford, General Motors, FIAT, Hyundai, and Toyota in order to try to estimate the additional fees and profit charged by the average car manufacturer. Not wanting to address the additional fees associated with the distributor (no idea who to trust there), I found a website that has the Invoice Cost listed so that it could be used instead of MSRP for comparison to the template’s Amortized Unit Production Cost. (Advice: Look at the invoice cost when you buy, not just the MSRP)
I tried to validate this template by proportionally distributing the weight of the baseline models of some of the most popular cars in America and using the yearly sales figure of the respective model to address learning curve. The results were the following:
As you can see, I’m getting pretty good results for production costs with this first draft, and the initial goals have been achieved. Due to the research I conducted for this template, the team has the following takeaways:
I plan to increase research into commercial automotive vehicles as a side project. I want to expand the scope of the project at some point to include electric vehicles, hybrid vehicles, and the millions of lines of code that can be found in modern cars. Hopefully, I’ll be able to assist in creating a more complex model that more so fits convention in the automotive industry. I will keep you guys updated if I uncover anything useful.