What was missing in the MRP for the APS to emerge
Many people ask us what APS software is, what it does and what kind of demand it is relevant. So in this article we will talk a little more about how the APS software and the main gaps in the MRP and ERP systems that it supply.
The origin
Production planning, in its broader concept, has always been in the heart of industry. It may be that in recent decades it has not been treated with the consideration due to other areas, but it has the role of protagonist in the history of technology for management in the industry. In the 1950s, computational development has introduced the computer-aid design, but its high cost made it accessible for a few, in case the aerospace and automotive industries, which have) very high technical demand. But the management was still "in the arm". The growing economic expansion, along with industrial development, has made planning challenges more critical. Thus, the first business systems initiative for the industry was neither a financial nor commercial system, but it was MRP (Material Requirements Planning), created in the 1960s by Oliver Wight and Joseph Orlicicky, precisely to plan materials to manufacture and buy.
For those of you who are already familiar with the subject, MRP basically generates net requirements for these materials based on finished product requirements and the explosion of mesacross the different levels of a product structure, generating the requirements for each material at each level, whether an intermediate component produced internally or a purchased raw material. Each of these requirements has a date stipulated by leadmes resupply standards for each item. MRP is nothing more than a series of simple, interconnected calculations, but in industries with many components and levels of structure, this is a real help, especially considering the period in which it emerged. Since innovation loves problems, the growth in demand for manufactured products after the 1960s faced the next planning challenge: capacity. It wasn't enough to simply know which materials would be needed in each period, but also whether it would be possible to produce or purchase the indicated volume, as well as to understand the financial impacts. Thus, in the 1980s, MRPII emerged. In it, resource management began to be considered, with an interface with the financial and engineering areas to understand the physical and financial needs of labor, machinery and materials for the execution of a production plan.
MRPII defines the MesProduction Plan (MPS), which defines the finished products to be produced in a given period (already broken down by SKU). Therefore, knowing the finished product requirements and roughly understanding what is possible to produce, the requirements for all materials are calculated using the well-known MRP. Finally, all requirements generated by MRP are validated. It covers all operations recorded in the product's Manufacturing Schedule and consumes the time to produce those products from the full capacity of each production center. This is essentially what many know and practice as machine load in industries. These concepts weren't all ready-made with the birth of MRPII; they were refined over time. Growing industrial demand and this technological evolution spreading to various areas still lacking in technology led to the emergence of Integrated Management Systems, or ERP (Enterprise Resources Planning) in the mid-1990s. These integrate the different areas of a company: accounting, human resources, finance, engineering, manufacturing, sales, among others. ERPs emerged seeking to incorporate one or more of the functions described above. The limitations of existing systems to ensure effective production scheduling encouraged the creation, in the mes1990s, of the concept of Finite Capacity Scheduling (FCS), which later evolved into Advanced Planning and Scheduling (APS). It is vital to understand that both systems emerged to address the deficiencies of other systems in managing finite production capacity, as well as queue management, routing, and synchronization between production operations. Let's explore these limitations.
:Fixed Lead Time
You, like the vast majority of people worldwide, probably commute daily from home to work. Although you may have a specific time frame for how long your commute typically is, if you monitor this time daily, you'll notice significant variations, especially if you live in a large city. This time can fluctuate even more if there's construction or an event on a street along your route that forces you to change your route. A factory is a megalopolis with more or less congested routes every day. For you, leaving 10 minutes early to ensure you arrive on time may cost little. However, for a factory, these "10 minutes" on the scale of hundreds or thousands of Production Orders, deadlines, and resources generates a very costly error from a planning perspective. Both MRP and MRPII use fixed leadmes , meaning they don't consider that your commute to work will change if you change your route or if the roads are congested. The more operations per product, the more resource alternatives or routing options, and the greater the demand variation over time, the greater this divergence will be. The direct consequence is the difficulty industries have in determining delivery times and ensuring healthy inventory levels. It's no surprise that most companies still rely on these systems and, not coincidentally, they represent a growing number of industries complaining about delays, shortages, and overstocks.
Batch or batch processes
Another point of attention is batch or batch processes, such as thermal treatments, painting booths, galvanization, dyeing, wear outlaws, among other types of operations that process multiple products simultaneously. MRP and MRPII do not distinguish these processes from those that are defined through a time by item (or weight, or length) or rate per hour. Thus, the load of these processes is poorly sized and the defined production plan will be less likely to be performed properly. Some claim that these processes are not a bottleneck in their industries always. This tends to occur in the thermal treatments of metal-mechanical industries, but in virtually all other cases these processes have a strong tendency to become bottlenecks quite often. In addition, based on a few hundred industries we have visited in many years, we can say that approximately half of them have this type of process, thus having this problem.
Synchronism
When there is more than one operation to transform a product, there is naturally a sequence to follow between them. It's impossible to package the product before painting or assembling it. The Manufacturing Schedule is the reference for this order between activities. However, the CRP doesn't handle this synchronization aspect well. Typically, this is handled in two ways. First, it loads all operations into the mescapacity period. However, depending on the queue that forms in each process, a subsequent process would be allocated to the next period, and its expected delivery date would be changed, resulting in a delay in this Production Order. In other cases, especially when process times are longer or there are many operations, one set of operations is defined as being in period 1, while another is in period 2, and so on. This creates a sequential logic over time, which is good, but the usual consequence of this scheduling is a significant dilution of production. Total production leadmes increase, the time to add real value to this total becomes shorter, and work-in-process inventories consequently also rise. This second scenario tends to generate fewer delays, but at the expense of creating inefficiencies in the system.
Finite capacity and restrictions
Perhaps this is the main factor that makes the result of MRP just a feedback of a test that cannot be required minimum grade because it is not known if it will be possible to solve it. Defining what you should produce or buy without validating capacity in theory would not allow us to charge results consistent with the factory. Usually there is not only one bucket with a capacity that is filled either: it is not just a machine or a job that restricts capacity. People who operate the machines, the tools used, the physical space between sectors and various other criteria are usually restrictive. A specific capacity analysis for the main restriction of each process or, worse than that, just for the bottleneck, is very limited. Now, if we join this finite capacity factor with its restrictions on what we talk earlier - the synchronism - we will see that the hole is below. It is no use making a machine load to limit capacity if we do not see that that product, which is apparently promised to be produced and delivered, can no longer be processed because today an indispensable tool for its production is being used in another machine. This kind of situation, involving tools and labor, is more frequent than you think.
The sequence
If we analyze Taiichi Ohno's seven Lean Manufacturing losses, we see that two of them are directly related to production sequencing: waiting and processing losses, the latter through setups. Some processes have setup times that can reach 50% of the total available time at certain times. Ask experts how long it takes to change a loom cylinder or wash a tank before producing allergenic foods. Depending on the sequence of operations you define for each resource, you can significantly reduce mold and tool adjustment times, tank and pipe cleaning, color adjustments, and even operator movement. Because these systems (MRPII/CRP) don't consider the sequence of operations, they tend to use average setup times, which can range from 5 minutes to 3 mes, and using an average will likely cause problems in process completion estimates. Since there is usually more than one operation to produce something, this error, which may seem small, is reflected in subsequent processes, and a domino effect occurs, distorting the initial planning. Along meslines, if we don't dictate a sequence, a shift might produce a specific amount and leave products for the next shift that can't be processed by mesperson because there isn't enough staff to perform the setup on that particular shift, or because the product requires more operators per machine and there aren't enough people on that shift, among other possible factors. This will generate longer waiting times between processes, another loss for Ohno. Ultimately, in practice, we see that systems that ignore important factors for assertiveness encourage a vicious cycle of empirical, oversimplified goals and controls that hinder the connection between actions and results. To eliminate these identified gaps, it's necessary to work with a more advanced method and technology. This is where APS comes in.
But APS means Advanced Planning and Scheduling, ie it is software for advanced production planning programming. APS is an expert system, its idea is not to replace the ERP, but to complement and meet its gaps, working on it, processing production and sequencing information, taking into account all restrictions, finite production capacity and the necessary synchronization, thus showing the best way to perform production according to each company's strategies. Neo is the largest APS consultancy in Latin America. Realize the sequencing of its production with those who specialize in technological solutions.