By: John Maher | January 20, 2025 | 7 min read
In my last post, I talked about the challenges of using spreadsheets to manage production planning and scheduling in complex manufacturing. Some of you have probably experienced those challenges first-hand, but if you want to read that post, you can access it here. If you’re ready to move beyond manual processes, today I will explore APS software further and dig deeper into algorithms, the logic behind Advanced Planning and Scheduling (APS) software.
What is an Algorithm?
In its broadest sense, an algorithm is a set of rules or a step-by-step procedure for performing a task or solving a problem. Algorithms are fundamental to all aspects of computer science from artificial intelligence, databases, and encryption to search engines, data compression, and optimization problems. It’s their ability to solve optimization and synchronizations problems that makes them so critical for advanced production planning and scheduling.
The Difference Between Algorithms and Formulas
When you talk to colleagues about replacing your production planning spreadsheets with APS software, someone will likely ask: “Can’t you just do that in Excel?” (If they don’t ask, they’re probably thinking it.) Before I answer that question, though, it’s helpful to understand the difference between an algorithm and a formula.
Although they can be very sophisticated, formulas are essentially equations designed to calculate a value. Algorithms can encompass formulas, but they also include logical structures such as loops and conditional statements, which are not part of formulas. Algorithms are also designed to outline the entire process of reaching a solution, including decision-making and iterative computation.
You can create algorithms in Excel, but anything that rises to the sophistication of a production scheduling function usually requires using Excel’s Visual Basic for Applications (VBA) programming environment. VBA expertise may be a handy skill to have, but creating your own algorithms means taking on the task of continually updating and debugging the algorithms as the business changes.
With off-the-shelf software, hundreds or even thousands of customers are funding the vendor’s investment in the resources necessary to keep the application up-to-date and error-free. As we discussed in my last blog post, APS software also makes adapting to unforeseen events, like a new order or unexpected bottleneck, a lot faster and easier as the vendor builds adaptability into the application. Plus, you don’t have to worry about your one VBA expert leaving the organization. Reputable software companies like Synchrono leverage proactive succession planning and adhere to enterprise software development best practices, ensuring continuity and minimizing intellectual property risks.
A Deeper Dive into APS Algorithms
The type of algorithm required is largely based on the complexity of the flow or routing in your manufacturing environment. If the flow is relatively simple, e.g., a limited number of process steps or a similar flow for all products in the value stream, then optimization techniques such as Linear Programming, Genetic Algorithms, and Constraint Programming are appropriate.
Manufacturing environments with more complex routing, many dependent resources, and highly variable flows from product to product require an alternative approach. Heuristic algorithms are a good example. Without going too deeply into the topic, heuristic algorithms are designed to find good enough solutions to a complex problem when finding the perfect (or at least optimal) solution would be too resource intensive. In other words, heuristic algorithms and hybrid variations thereof are often used when traditional optimization would take too long to provide an optimized answer.
Heuristic algorithms also address the challenge of convergence. That is, the more complex the flow, the more variables and, thereby, possible solutions. This forces the algorithm to end further from convergence on the optimal solution. Optimization algorithms that don’t converge on the optimal solution can vary widely from run to run without any of the data changing. As every production planner can attest, having a very good schedule that is also stable is far more critical in a complex manufacturing environment than creating optimized solutions that are highly variable from run to run.
Furthermore, adding in convergence points, such as welding or assembly where multiple items are combined, requires a heavier focus on the synchronization of parts than optimization of individual resources. Without synchronization, a lot of time is wasted at convergence points waiting for parts that were not optimized for upstream resources to run when they were needed.
Why Not Take the Proven Path?
Obviously, we’ve just scratched the surface when it comes to discussing how algorithms are used to synchronize resources and optimize production schedules. I’m sure we have readers who are hungry for more details, but a lot of you are probably wondering whether you can just skip the lecture series on this topic and still benefit from APS.
The good news is, you can. Our APS solution, SyncManufacturing®, is a patented APS application designed for complex manufacturing that utilizes many different types of algorithms to optimize production scheduling. During implementation, we’ll work with you to make sure the system applies the most effective algorithms to derive stable, workable production schedules that help you meet your customer commitments while maintaining the agility needed to respond to whatever comes next. No programming of algorithms needed!
To learn more, reach out to us or schedule a demo.
