Tag: forecast

ERP is an Oxymoron

Have you ever had one of those moments when a thought hits you that is so obvious that you wonder why it never occurred to you before? I just decided that there couldn’t be a technology less aptly named than Enterprise Resource Planning, more commonly known as ERP.

Before I go into why, let me lay the groundwork so we’re all using the same terminology.

What is ERP?

From a manufacturer’s perspective, ERP was born with the addition of MRPII to the core financial functionality already found in existing systems. Pretty much all of these early systems provided modules for GL, AR, and AP. (Sometimes called “GLAPAR” in industry jargon.) Many of the early ERP solutions provided some procurement functionality as well as sales orders, though not many offered manufacturing-oriented functionality like the ability to configure products on the fly.

Sometime in the 80s, ERP vendors realized if they wanted to target the manufacturing sector, they needed to provide tools to help manufacturers deal with what is arguably their biggest issue: inventory. That’s when Material Requirements Planning (MRP) – a concept that had been around since the 50s – was codified in software form.

There is a belief that manufacturers are slow to adopt technology. I think it’s more that technology is slow to adopt manufacturing. Historically, ERP vendors have focused on manufacturing last, and many never quite get it right.

And that leads me back to my original premise: ERP is an oxymoron.

Garbage In/Garbage Out

As Trey Jordan recently wrote in his post Going Lean: Should You Replace Your ERP System? (insert link), ERP systems are great at collecting financial transactions from all over the enterprise. The E in ERP is just fine. My problem is with Resource and Planning.

In this post, I’m just going to look at inventory, since many ERP systems don’t even try to manage capacity. If you want to look at this from the capacity perspective, you might enjoy our white paper: The Next Generation of Planning and Scheduling Solutions.

The core resource planning tool in ERP systems is, of course, MRP. Put simply, MRP looks at future material requirements and works backward using inputs such as current inventory levels, order lead time, and so on to create time-based requirements for raw materials and components.

In theory, that makes perfect sense. If I want to make 100 widgets by the end of the month and it takes me two weeks to do it, I need to make sure I have the materials available by mid-month in order to reach my goal. If some of the materials required are subassemblies built in-house, then the system issues production orders so they will be available when needed as well.

The problem lies not in the algorithms, but in the inputs. Or to use a technology axiom that’s been around even longer than MRP: Garbage In/Garbage Out.

Where does the order for 100 widgets come from? The quick answer is the production schedule, but what are the inputs into that? In push-based manufacturing environments, the primary input into the production schedule is the sales forecast.

Therein lies the problem.

Sales forecasts are always inaccurate; the only question is by how much. I just Googled “sales forecast accuracy” and got 2 million hits, most of them having to do with why sales forecasts are always wrong. I skimmed a few of the results to see how accurate forecasts are on average, and the answer seems to be around 75%.

However, many of these articles and posts looked at forecasts from the perspective of the sales team: Did sales hit the numbers? A 100% accurate forecast from the perspective of the VP of Sales can still be wildly inaccurate from the perspective of production if the products sold were different than those forecasted.

Wisely, many manufacturers have learned not to trust the forecasts generated by sales. They hold weekly S&OP meetings to look sales leaders in the eye and review what’s in the forecast so they can adjust their production schedules based on what seems realistic and doable. These meetings can get tense. And, as much as the participants would like to apply proven processes to their S&OP meetings, they don’t do much to fix the core problem: Everyone is still guessing.

Manufacturing Demands a Different Approach

Maybe I should cut the ERP vendors a break. It’s not that their systems are coded poorly; it’s that they start with the wrong premise. That is, they were developed for push-based manufacturing, a generally accepted practice even though MRP has never been proven to provide the sustainable performance improvements manufacturers need.

The opposite of push-manufacturing is pull, where resources (including capacity) are synchronized to customer demand. Pull is a core element of manufacturing principles such as Lean Manufacturing and JIT Inventory Management, so many readers are probably familiar with it. We often used pull-manufacturing or pull-replenishment synonymously with Demand-Driven Manufacturing, though our software Platform also applies other core principles of Lean such as Theory of Constraints.

If you’re tired of dealing with an Enterprise Resource Planning application that does nothing to help you plan resources effectively, I encourage you to investigate Demand-Driven Manufacturing. Our website is filled with white papers, articles, on-demand product demonstrations, and case studies that can help you learn more. If you’re brand new to the concept, check out our YouTube video: What is Demand-Driven Manufacturing?

June 18, 2018
It’s Time: Manufacturers Need to Cut Ties with MRP and Spreadsheets

A few years ago, Aberdeen Group did a study that showed that 63% of “best in class” manufacturers still used spreadsheets for planning. With percentages this high, it’s probably safe to say that there is a lack of trust in planning tools like ERP and MRP even in the best-run companies.

Of course, spreadsheets come with their own set of issues. You may have a certain amount of confidence in the spreadsheets you’ve created, but what about those from your colleagues? Do you know what formulas they use to arrive at their conclusions? Where did they get their data? If the creator of the spreadsheet goes on vacation or worse – leaves the company – how quickly could a new planner take over their role using the existing planning tools and methodologies?

Spreadsheets also contain computational and data errors. Forbes published an article quoting “various studies” that put the rate of significant errors at 88%. Though no specific studies were named, you probably don’t need that extra level of validation. You know the spreadsheets you use have errors in them. Otherwise, why would you still be having such a hard time synchronizing inventory and production to demand?

The Problem Lies in the Basis of the Plan

Actually, spreadsheets are pretty powerful tools, even with the occasional error that creeps in. Likewise, MRP and ERP usually do exactly what they are supposed to do. They create time-phased material requirements and production plans based on the parameters entered into the system: forecasts, reorder points, capacity and so on. The calculations are so basic that even the most rudimentary software applications get them right.

The problem lies in the basis for the plans themselves. At the root of all production and material requirements plans lies the forecast, often generated by the sales or marketing department. Even when calculations are based on an analysis of historical data, it’s still a forecast. And, as we are all painfully aware, there is no such thing as an “accurate forecast.”

Not trusting what they’ve been handed, inventory and production planners use spreadsheets to massage the forecast data before it is entered into the system. Some of these comments might sound familiar:

“We always see a bump in demand for these items in June, so we need to increase production now.”

“I know sales has a quota on this new product, but I think they’re being overly optimistic. If we cut the requirements by about 10%, we can deal with it later if they actually manage to reach their quota.”

“I read in the news that there may be shortages of this material. Let’s order extra now so we can stay ahead of the problem.”

At the end of the day, the forecast is still a forecast, even with the wisdom of inventory and production planners added in. Unfortunately, the new forecast may not represent reality any better than the original forecast received from sales, and ERP and MRP have no choice but to translate those erroneous assumptions into equally flawed material requirements and production plans. In turn, these flawed plans translate into all sorts of issues: expediting orders, late nights, increased overtime, missed deadlines, angry salespeople, angrier customers, inflated inventory levels, blown budgets, tense meetings in the executive conference room and bland food diets to prevent your indigestion from turning into a full-blown ulcer.

Grounding Planning in Reality

Since there is no such thing as a crystal ball that shows future demand, manufacturers who want to break free from this vicious cycle need to replace forecast-based planning with reality-based planning. In short, ditch ERP, MRP and spreadsheets (at least for replenishment and production planning) once and for all.

The only way to do that is to synchronize production and material requirements to actual demand and supply as well as what is happening on the factory floor. It’s called Demand-Driven Manufacturing, and here’s a quick definition:

Demand–Driven Manufacturing is a method of manufacturing where production is based on actual customer orders (demand) rather than a forecast. This process is accelerated by technology that automates, digitizes data and connects every function within the demand-driven organization and to every layer of the supply chain.

We created an entire platform called the Synchrono® Demand-Driven Manufacturing Platform that sits on top of your current ERP system and synchronizes all elements of production to demand and supply. There is no need to rip out your current ERP or MRP applications; our tools use actual customer demand, supply status and the reality of the factory floor to synchronize production. The methods used, such as eKanban, Lean Six Sigma, Theory of Constraints, are no doubt familiar to you.

If this is the first time you’ve taken a close look at Demand-Driven Manufacturing, we have several resources which can help you build a solid foundation for discussions with others in your organization:

White paper: The Next Generation of Planning and Scheduling Solutions

White paper: How Technology Will Connect Your Enterprise and Create the Demand-Driven Manufacturing Factory of the Future – Today

White paper: Why Become More Demand-Driven? Responding to Customer Needs

We also produce a YouTube channel where you can access several educational podcasts and videos that explain some of these concepts in more depth.

As always, if you have questions, please add them in the comments below, or reach out to us directly. We would love to hear from you!

March 19, 2018
Weathering the Next Recession

The global economic recession that started in 2007/2008 hit the manufacturing sector hard. In February of 2009, The Economist even published a piece called The Collapse of Manufacturing. If you recall, this was right about the time the automotive manufacturer bailouts kicked in because they, like the larger financial institutions, were deemed too important to the US economy to fail. It felt like we were on the verge of an economic collapse.

Whether they supported that bailout or not, most people recognized that it was a short-term solution to a long-term problem. The auto industry, like many other manufacturing sectors, was awash in excess capacity. When consumer demand dropped due to the recession, this excess capacity could no longer be ignored because it led to mountains of unsold inventory and raw materials sitting in the supply chain.

Many companies didn’t survive the recession, at least not in their current state. In 2010, Harvard Business Review said that 17% of the companies they were following in a post-recession study had gone bankrupt, were acquired or went private. Opinion makers and pundits differ as to whether the recession actually ended in 2010, so that finding might have been a bit premature.

Things are looking up

While there are still plenty of naysayers who can find the cloud surrounding every silver lining, many economists agree that indicators are improving as we approach the end of 2017. The MAPI Foundation predicts a pretty substantial growth in manufacturing output at least through 2020.

As I write this, unemployment has fallen to 4.1%, a 17-year low, with manufacturing being one of the primary beneficiaries. According to MarketWatch, so far in 2017, manufacturing jobs have increased by 138,000 after falling by 34,000 in the first 10 months of 2016.

Learning lessons from the past

Let’s assume (and hope) these predictions are accurate and the positive trends continue. Now is not the time to get complacent. Unfortunately, recessions are like earthquakes. You know they will happen. You can even monitor for signs of an impending quake. But it is almost impossible to predict when and where they will strike and how much damage they will do. The MAPI Foundation’s growth in output levels off in 2020 because they are allowing for the possibility of another recession. They don’t go so far as to predict one, but they know it will eventually happen.

Clearly, pulling back and not investing in anything isn’t the answer. Instead, the key to surviving the next recession is to avoid making the same mistakes. Demand-Driven Manufacturing can help.

Understand your true capacity – One of the key principles behind Demand-Driven Manufacturing is the Theory of Constraints (TOC). In TOC, there is always one resource that limits the factory’s ability to meet demand. For example, it might be a work station that has a constant backlog, impacting all downstream operations. In Demand-Driven Manufacturing, that constraint becomes the pacemaker for production, keeping work flowing at a constant rate, or load (e.g., CONLOAD) Watch this short demo of CONLOAD to see what that looks like.

Many manufacturers using CONLOAD (part of SyncManufacturing) find that this process actually helps them to increase capacity; because constraints are managed, work continuously flows.

Avoiding excess inventory – In Demand-Driven Manufacturing, production scheduling is based on actual orders; nothing is produced until it is needed. For example, a manufacturer may believe there is going to be an uptick in orders for a particular widget. They might even add capacity in terms of more modern equipment or higher-skilled operators. But that widget will not be produced until an order is actually received. Likewise, the components that go into the final product will not be produced either, lowering both finished goods and WIP inventory.

December 4, 2017

“test”