What are the 5 process types?

Download this article in .PDF format
This file type includes high resolution graphics and schematics when applicable.

It may be surprising to learn that many engineers with great talent and a depth of experience have a hard time answering questions about the type(s) of manufacturing environments that exist in their company.

Most manufacturing environments fit into one of five general categories. Repetitive, Discrete, Job Shop, Process (batch), and Process (continuous).

Most companies use more than one of these environments to get a single product out the door. This is certainly true considering today's use of the supply base versus the historical practices of vertically integrated companies. Vertically integrated companies often had all five environments.

Here’s a closer look at the five environments:

Repetitive: This category, with some exceptions, is best described as having dedicated production lines that turn out the same item, or a closely related family, 24/7 all year long. Speeding up or slowing down the speed of the operation modulates differences in customer demand. There is little setup and changeover activity. If the peak line speed cannot keep up with demand, a second line is added. If demand is not enough for a dedicated second line, it is met by a second line operating in Discrete mode that also makes other products.

Discrete: This environment is highly diverse. It covers a range from few setups and changeovers to frequent setups and changeovers. The products being made may be alike or highly disparate. The more unlike the products are, the longer is the unproductive set-up and tear-down time.

Job Shop: Job shops rarely have production lines, they have production areas. The area may assemble only one version of a product, a dozen versions, or even a couple dozen. If demand grows, the operation is turned into a discrete line and selected labor operations are replaced by automated equipment.

These three environments, taken together, are a continuum for mechanical, electromechanical, electronic, and software-driven hardware products. At one end, manufacturing is continuous. At the other end, it is highly intermittent. Designers at the continuous end must be versed not only in the product’s design, but also the design of the process equipment. The more repetitive production is, the more likely the environment is dominated automated equipment. Production personnel rarely touch the product; their role is to oversee the equipment and assure it functions properly.

Designers at the Discrete end, with Job Shop being the ultimate case of Discrete, must be versed in product design and know when to involve equipment instead of labor for production. Designers in Job Shop environments generally cannot justify automated production equipment and must be expert in parsing the design so that components and subassemblies are made or acquired economically. They must also be expert in understanding how their design can be manually assembled.

The remaining two environments have many analogies to those just described. The primary difference is that scientists and chemical engineers typically "design" the product, and the process equipment is handled by engineers versed in repetitive and/or discrete design. The product is usually described by a formula plus a bill of materials.

Process (continuous): These operations are analogous to Repetitive; they run 24/7 all the time. The primary difference is that production materials are gases, liquids, powders, or slurries. In some cases, such as mining, they can be granular or chunky materials. Design considerations are analogous, except the disciplines to create final product and production process are more diverse.

Process (batch): Batch-process operations are analogous to Discrete and Job Shop. Sometimes one batch is all it takes to meet demand. Sometimes it takes several batches. The equipment is then cleaned and the next product run. In some cases, Batch processes can be continuous in nature, making one batch after another of the same product. This is common when the composition of raw materials cannot be made to a strict standard. Each batch must be analyzed and some change made to the original formulation to turn out a final product that meets the spec. Again, design considerations are analogous and the disciplines are more diverse.

Find out how 3D printing has become the sixth manufacturing process in my MD Aug 2018 column.

See the 5 Types graphic upgraded to the 6 Types graphic in my MD Sept 2018 column.

In order to continue enjoying our site, we ask that you confirm your identity as a human. Thank you very much for your cooperation.

17/06/2021

Axel Vanhooren

Download this Article as PDF

Importance

We can observe different types of processes. Being aware of how they differ from each other is important in order to deal with them in a suitable manner.

Content:

1. Structured Process
2. Case-type Process
3. Research Process
4. Engineering Process
5. Artistic Process
6. Conclusion for the Software Engineering Process

1     Structured Process (Production Process)

Structured processes can be production processes producing products and services. The process of an assembly line is a good example. Not all production processes are structured, but there is a tendency to make them as structured as possible.

They are executed as a series of consecutive well-defined discrete steps. Each step produces a particular result. These processes are or can be repeated over and over again. They have to produce the same product or service, even if they are customised. These processes have to be executed exactly the same way. They have to be executed strictly. They can be formalised. People executing those processes don’t need to think a lot. They can simply follow the recipe.

We are able to create the right - at least to some degree - environment and conditions in which the process is executed. The environment, the conditions, the resources and the input are under control. They are (rather) stable and reliable.

These processes are predictable in resources, duration, output and quality. They are easy to plan.

Control can be increased by adding control points, verifications and tests at some points in the process.

2     Case-type Process (Semi-structured, loosely structured)

Although some processes produce a same or very similar result, their execution might be unique. Nevertheless, we can discern global phases in the process. These phases may have clear and well-defined boundaries. These are defined by exit conditions and prerequisites. And the phases may even overlap.

The process may have to deal with a large variety of situations in the real world. It has to continuously adapt to it. The process can be dependent of various inputs of its environment. And the system or person executing the process has no control over when and what input they will receive. External actors who behave unpredictably are involved. They are not under control of the system.

Each phase contains processes, parts of processes or even single activities. And some activities can even be created ad hoc. Each execution of a phase is unique. Nevertheless, there are similarities with the execution of the same phase in other processes.

The entire process or even phase cannot be formalised. It can be described. The entire process is defined when the phase is executed accordingly to the specific situation of the case at that moment. Indeed, the people executing the phase have to decide what processes and activities are necessary and which can be executed when. They are more autonomous and self-guiding. They are the pilot. Returning to an earlier stage is not intended, but ‘the pilot(s)’, may decide to do so if necessary.

Some of the individual activities or smaller processes can possibly be formalised. Phases, parts of processes, methods and outputs can be planned (not necessarily with time estimates).

Often, it requires an expertise to be able to decide what actions to be taken and what smaller processes to trigger. Obviously, such processes cannot be managed and estimated like structured production processes.

A formalised process can be suggested as possible path for processes which are semi-structured by nature. But since a rigorous execution of a strictly formalised process would too often lead to failure, they cannot be imposed.

Software methodologies and software development processes are good examples of a case-based process.

3     Research Process

Research is meant to acquire knowledge and understanding. In particular, it works through the verification of hypothesis. The observation of what is happening during experiments and their results provide if not answers, at least more information or confirmation.

The research process doesn’t create something new. It is meant to learn from something that exists. It is not an engineering method, although it can embed and use engineering methods to create new experiments. Or, it can be embedded in an engineering approach.

The Scientific Method is a method that focuses mainly on the execution of a single experiment. In this case, we consider a whole set of experiments conducted with a common goal.

Basically, a goal or hypothesis is established. Various experiments are needed to study and clarify the question. Each experiment provides information. This will be used to decide if and what next experiments must be done. Some future experiments may be known and planned. But other experiments to be conceived after those may well not be known at all. 

A single experiment can be planned. All the experiments intended to be performed can be planned. But beyond that the process is unknown and unpredictable.

Only the short term can be planned and is predictable in terms of time, materials, resources and skills. Beyond, the next range of intended experiments, nothing has been decided.

4     Engineering Process

The Engineering process aims to create something. The main idea is to follow a methodical approach. It is characterised by three main activities: Investigation and Research, Thinking and Experimentation (Prototype, simulation, trial-and-error, ...). 

Investigation and research can be planned if the information sources are well-known and up-to-date. This is rarely the case. We may not even know what information will be needed. We may depend of decisions and of information of outside sources. Information can be acquired through collaboration. Often this activity is hard to plan and unpredictable.

Intellectual activities are unpredictable. A lot may happen; such as confusion, underestimation or forgetfulness; that hinders the mind. It’s easy to disturb the mind. Or we may be upset or tired. Our mind doesn’t always think efficiently, effectively or at a constant pace.

Either the end-product is used to gain feedback or separate experiments, such as a prototype, are used. Experiments provide information, to some degree, about the behaviour and about the suitability. It doesn’t explain how things work.

The more the experimental approach is used, for example for shaping and transmuting the solution instead of for specific questions or for adjustments, will make the process more unpredictable.

Engineering processes often include a high dose of collaboration making the process even lesser predictable.

Within engineering, Systems Engineering is commonly confused with Product Design.

The goal of Systems Engineering is to design a system that is able to function, to execute work, to produce goods or services within its environment. Note that systems and their supra-systems and environment are intrinsically linked. Systemic Thinking and Systems Engineering form a perfect match.

Product Design allows the creation of a new product to be sold to customers and used by consumers. Design Thinking applies well to this context. Product design can, for example, be used for designing the user interface of systems.

Both purpose and processes are very different from each other.

5     Artistic Process

A pure creative process looks more like this. It may contain research and exploration and rely on insights and ideas popping up in the mind. Such processes cannot be formalised. It is impossible to forecast when the insight or when a brilliant idea will emerge. The incubation period may last days, weeks and even year. We have no idea when the inspiration or moment of illumination will be.

Creative processes are very unpredictable.

6     Conclusion for the Software Engineering Process

In order to defining, execute and manage a process in a sensible way, we have to be aware of the type of approach or process we deal with. The control over a structured process can be increased by formalising it, adding control points, establishing performance indicators, setting deadlines, logging activities and resource usage. Such measures won’t make sense at all, for example, on the artistic process. It would kill the benefits of the artistic process.

Each type of process requires an environment meeting its own specific prerequisites. Unfortunately, too often processes seem too often to be considered as structured processes, either by lack of differentiation or because of the urge to know and to be in control. The needed conditions required to execute the process in the best circumstances are then ignored.

It is possible to embed different process types one into another. For example, a marketing initiative can be set up as a case-based process. It will make use of an advertisement which is an artistic product.

There are many types of software systems and applications. Software systems allowing managing a nuclear power plant is “slightly” different from social media features. Even within a single system, most important criteria applying to the core of the system are different from those for the user interface.

Software Engineering is certainly not a common structured process, neither is it a research process. It relates more to the case-based process and to the engineering process.

Axel Vanhooren

Freelance Consultant - IT Re-Thinker - Business Analyst

Download this Article as PDF

OTHER PUBLICATIONS

The Analyst’s Red Book

A Visual PDF-Guide about Enterprise Informatics Empowering Analysts and Architects

Download (Latest finished version @ http://analystsredbook.net/)

Download (Latest draft version; “as is”; with more content; maybe only temporarily available)

ARTICLES

Problem Solving: Concepts and Approaches for Systems and Strategies

Systems Fundamentals to Empower the Systems Analysists (Systems Thinking)

Rediscovering the Purpose of Systems Analysis (short version)

25 Ways to Disempower Your Business Analyst - STOP Killing the Analyst