At the core of a supply chain digital twin there needs to be a representation of the manufacturing process. Multiple complexities may need to be factored into this model, such as production switching costs, utilities, minimum run sizes, and so on. Modeling becomes even more challenging when you factor in other production or tolling sites, as well as the dependencies across sites. As you extend backwards from production into suppliers, there are aspects that should be modelled here as well including different purchase minimums, costs, and lead times varying by supplier.
Finally, you have the downstream supply chain consisting of warehouses, distribution centres and customer shipto locations. Things get complicated when you try to factor in duties and tariffs or product substitution options. As you can imagine, it can be very challenging to model these interrelatedelements in a spreadsheet-rather than using a solution designed specifically for that purpose.
The other big limitation of a spreadsheet is that it wasn’t designed to do mathematical optimization at scale to solve real-world problems-taking into consideration anywhere from tens of thousands to millions of variables and constraints. Using a solution designed specifically to do mathematical optimization at scale such as Aspen Supply Chain Management (SCM) is extremely valuable because:
An optimizer will find the best answer automatically, whether the goal is to maximize profit or minimize costs across the end-to-end system.
An optimizer will recommend options that a person or business wouldn’t normally consider or didn’t know were even possible. That’s because it can easily deal with complexity in a way that a human mind cannot.
