Introduction

Before we get into talking about manufacturing methods, let’s go over some of the broader themes.  The first is an obvious question that the company must ask.  Should we manufacture this product?  Or should we get someone else to manufacture this product, and we buy it from them?  The former is called vertical integration, where the latter is called outsourcing.  You’ll probably heard in the last few years the concerns with outsourcing to other countries where manufacturing is cheap, such as China and Mexico, but a company can outsource within the same country – it just has to be a different company making the product.  Whether a company outsources or decides to make the product itself usually comes down to cost – whether the company has the expertise and equipment already, or whether it would be better to avoid huge capital investments and instead just pay someone else to do it.  One of the big advantages of outsourcing is that it transfers much of the risk to the other company.  What happens if all the demand for your product disappears and you’re left with costly equipment that you cannot use?  This brings to attention the concept of flexibility when it comes to manufacturing, and flexibility is key – to have the ability to react and respond to external changes, such as supply and demand, technological shifts, increased customization, and so on.  In today’s world, a company needs to be able to respond quickly in order to maintain a competitive advantage.  There are different flexibilities – operational flexibility refers to how you can change your actual operation as required.  For instance, the ability to reconfigure and reprogram your processes and equipment quickly in response to changing conditions.  Tactical flexibility refers more to a company’s ability to change to the mix and the volume of the products desired, as well as the ability to change the product for modification/customization.  Strategic flexibility refers to the ability to introduce new products.  Really, though, you don’t need to remember these individually, because they probably all more or less come to mind when you think manufacturing flexibility.  This is partly why 3D printing is becoming popular – the capital investment is low, and you have the ability to make more or less whatever you want.  Of course, there are severe limitations, such as the materials available and product volume, to name two obvious ones, but you get the idea.  3D printing is off the charts on the flexibility scale, and it just may be a disruptive technology in the manufacturing sector.

A Very Brief History

There’s significantly more that can be discussed on manufacturing in general before we get into the specifics, such as the rich history.  One particularly compelling example of the impact of manufacturing was the introduction of the Model T by Ford at the turn of the 20th century, when mass production really started to take hold.  It started with an engineer by the name of F. Taylor, who decided that production could be optimized by actually studying the processes and the capability of the workers (gasp!).  But the ideas were somewhat revolutionary, and in fact, some say that it was around this time that industrial engineering was born.  Taylor also suggested that tasks could be broken down, and that workers could become specialized in these repetitive tasks – which is beginning to sound a bit like a production line, isn’t it?  Taylor also got involved in planning out the layouts of factories to make everything more efficient and reduce the distance between processes.  H. Ford took these ideas and applied them to his factory, and the operators began completing repetitive tasks during shifts, and time became a focus – shaving even seconds off certain processes means that more cars could ultimately be built, and thus more revenue.  But this doesn’t sound very flexible does it?  Well, no, but he did arguably revolutionize manufacturing, so we can forgive him for it.  There’s the famous quote by him: “Any customer can have a car painted any colour that he wants so long as it is black.”  This began to get his company in trouble with it’s age old competitor, the very fine General Motors, who offered a whooping four models.  And so the idea of the flexible manufacturing paradigm began to grow in popularity.  Today, flexibility is a huge consideration – consumers have so much choice, and fickle taste and desires, and an unwillingness to budge on both lead time and quality, and the rise of globalization – all means that in many cases, a company needs to be able to turn on a dime.

Global Perspective

We can talk about how different countries have different approaches to manufacturing – yes, different countries.  Here’s are the basic plot synopses of three of the big ones:

• USA: historically excellent at research and development and innovation, crappy at turning this into commercial products that sell (with the massive exception of software), although this is changing.
• Germany: that german efficiency has lead to performance products of the highest quality.  Think of german made appliances and german made cars.  Highly skilled workforce.  Get it right the first time attitude.  Many company executives with Ph.Ds in engineering.  Less about innovation, more about improvement and perfection.
• Japan: good at taking innovative ideas and R&D and making useful products that do well commercially (so more or less the opposite of the USA): televisions, stereos, cars. Many countries intervened with trade action in order to limit the bleeding caused by Japanese manufacturing.  Who knows if the american auto industry would have survived in the 80s had it not been for the American government?  Or how it would look today?  But today the competition is catching up, and competitive advantages are being razed.

While we’ll save the discussion of design for another chapter, there is a design topic that’s very relevant for this section.  It’s part of the “Design for X” series.  You can replace the X with any number of relevant goals.  I’ll rattle some off: design for cost.  design for usability.  design for user-friendliness.  design for the environment.  design for aesthetics.  design for manufacturability.  design for assembly.  design for reliability.  It basically just means to consider the ‘X” in the component that you’re making.  It’s all about following guidelines to achieve the traits in the product that the X dictates.  For example: designing for the environment would mean that you consider how the product will be recycled, whether any materials will be used that are harmful for the environment, and so on.  Design for user-friendless is a big when your products are going to be used by consumers, who we have to assume don’t think like engineers, and the same things that are readily apparent and intuitive to engineers are likely not to the rest of the populous.  The iPhone comes to mind when you think of design for user-friendliness.

Manufacturability

Of course the one that we’re concerned about is design for manufacturability.  This topic seems like it would be a passing thought, and indeed, it’s not covered much in school (at least in my experience).  But if you ever take a job in industry where physical products are being made, and you’re design some new fangled fancy part, one of the first questions that you’ll be asked by your lead engineer is “how are you going to make that?”  Not you personally, of course, but how can we even make that part?  There’s a few major considerations.  Now, you might be able to physically make the part.  I mean, humans can do pretty much anything they put their mind to.  But can we make it easily?  Can we make it quickly?  Are the manufacturing engineers going to call and yell at us for making their job so difficult?  A bigger consideration is the cost.  How much will it cost to make that part?  Is it a simple process?  Then there’s the material that it’s made of: did you have a carbon fibre reinforced composite in mind when you made it lightweight and sexy looking?  Well, that’s going to be expensive and slow.  Even if you’ve been a good design engineer and considered these, and are making the part of boring old metal with a reasonable shape, did you follow more specific concerns? If the part is cast, do you have any sharp corners where local failures might occur?  Have you specified an undrillable hole?  An nonmachinable slot?  It’s not an easy task to design for manufacturability.  It requires skill and knowledge, and an imagination.  I say imagination because you sometimes you need to look at your part and imagine how it will be made – walk through the process.  You might catch something.  

Computer Simulations and Manufacturing Considerations

Sometimes, computer simulations will be run to determine if it can be made, so it’s not a trivial task.  For example: stamped components are used extensively in the automotive industry.  Take a piece of relatively high strength steel – usually with a yield strength between 340 and 700 MPa – with some thickness, usually between 2 and 6 mm – and cut out a the shape you want.  And then a series of hits from a massive press form the flat piece of steel into the shape you want, maybe for suspension link.  If there are abrupt form changes in the part, then you risk thinning the part so much that it is made weak, or worse – you end up tearing it.  You could also end up wrinkling it – sort of the opposite of tearing, because all the material bunches up together.  You also need to consider how the forming process is going to change the material properties of the part, because when you form it into the shape you want, you’ve yielded the material – making it stronger in certain areas.  If you need a part with a certain strength, which is often the case, then you’ll need to consider these ‘forming strains.’  On a side note, with the design of suspension components, car companies want an upper limit on how strong the arms are – in other words, they want them to break at a certain load.  For what possible reason?  If you hit a curb going fast, or get struck by another car, it’s desirable that the control arm buckles before transferring that load into the frame of the car, which is much more expensive and difficult to replace.  If the control arm breaks first, then hopefully the frame will remain largely unaffected.  So in this case, you need to consider the additional strength added to the arm by the manufacturing process.  A specific example, I know – but it goes to show that sometimes you have to consider manufacturing in somewhat unintuitive ways.  We’ll come back to design for manufacturing when we get into more detail about specific manufacturing processes, such as casting and forging.

Before we get into specifics…

So considering manufacturing when you’re designing a part is an important step.  It’s best to design the simplest part possible, which is easy to make and cheap to produce, with the cheapest material possible – you want adequate performance for the lowest cost in most engineering applications.  There are obviously exceptions, such as when you’re producing products that are more desirable and higher quality – like expensive cars – or when the quality and strength of the part matters hugely – like in aerospace applications.  But keep in mind, even in aerospace applications, they want to keep costs down, so the idea still applies.  Of course, their idea of adequate performance means something quite a bit different than if you’re designing a chair for a cheap furniture company – but companies don’t go around making parts more expensive then they need to be.

The majority of parts that you come across will either be forged, cast, rolling, extruded, made of of sheet metal, or billet machined.  Deciding which process to use really comes down to cost more than anything else.  Everything can more or less be traced back to cost.  If you’re making many parts, then it might make more sense to spend a bunch of money initial on dies and equipment, as the cost to make each part will become lower.  It also depends on the quality of part you’re looking for – what surface finish you want, how close the tolerances need to be, how strong, what material, etc.  But generally speaking, you’ll have a set of requirements for the part, and then you’ll choose whichever is the cheapest.  Sometimes the answer is clear.  If you’re making a stamped part out of thick sheet metal, then obviously that material will have been rolled (you still have to decide how to cut it however).  Large parts with complex geometry, like engine blocks, will probably be cast.  Parts that can be cast can likely be forged as well.  Usually when you begin designing a part you’ll have a pretty good idea of how you’re going to make it, and so at some point you’ll have to consider design for manufacturing.  Sometimes, it will be decided later on in the design process how the part will be made – you might switched from a billet machined component, to a cast component.  This may require design revisions and further analysis based on what properties you can expect from the chosen manufacturing process.  Don’t expect to be an expert on manufacturability of each part – it takes practice and intuition in many cases – but you should know the basics, and I hope that the next pages will provide them.