Systems engineering

Definition: Systems engineering is the field of engineering that deals with the manner in which complex and interdisciplinary engineering projects can be organized, designed, and managed. It focuses on coordination of separate teams, logistics, life-cycle maintenance, automation, and other ways of working with large projects. It does this through development of processes and design paradigms as well as interfacing parts of the project. Systems engineering often straddles the line between technical and human-centered disciplines, with business, organizational studies, and project management on one end and various disciplines of engineering on the other. Developed as a means to manage the massive projects of the Cold War-era Department of Defense (DOD) and the National Aeronautics and Space Administration (NASA) in the United States, systems-engineering paradigms have since become critical to diverse disciplines, particularly software development.

Basic Principles

Systems engineering developed from a need to coordinate large-scale projects involving not only a variety of engineering work but different organizations and groups as well. The term is first recorded as being used at Bell Laboratories for use on large projects, mostly work in support of the Allied war effort during World War II. Among other tasks, Bell statisticians helped in ammunition and materials sampling. The methodology was soon picked up by the DOD and NASA. The DOD found systems engineering useful because it allowed them to coordinate the new defense projects needed during the Cold War. Since secrecy was often important, those in charge found the ability to break a task down into components a useful part of systems engineering.

By determining the critical components and sequences for successful outcomes, those managing the projects found that they could increase the effectiveness of defense programs. Projects such as President Dwight D. Eisenhower’s interstate highway construction, the establishment of Strategic Air Command, and various intelligence programs all benefited from systems engineering because people with expertise in the technology could interface directly with those who needed the systems. They could then discuss the practical goals and desired outcomes of the project. The engineers could devise an abstract framework so as to meet goals. In addition, the DOD could find better ways to organize their own groups to meet emergent threats.

The American space program was also an ideal candidate for systems engineering. Not only the rockets but also the infrastructure and methodology were designed from the ground up. After President John F. Kennedy’s 1961 declaration of intent to go to the moon, the space program became dedicated to producing a moon-capable system. Thus NASA designed a process that would develop the technology in a timely and efficient manner by determining what components were likely critical and devising a way to test as many of these features as possible.

Systems engineering has spread across many fields since the 1960s, and in the twenty-first century, many projects begin by outlining goals and developing systems and teams to accomplish them. This is a particularly common practice in software development. Systems engineers view their problems holistically and look at how the parts fit together, then divide the workload into manageable chunks with clear goals. This is accompanied by analyzing the project to find which steps depend on others and prioritizing those on which others depend in what is called the critical path.

Core Concepts

Abstraction. In order to prioritize and decide upon its goals, a systems engineer must be able to look at the abstract of the situation. For example, a systems engineer must be able to see commonalities between different portions and tasks so as to group them. Also, the ability to temporarily ignore unimportant details and view the project as many interrelated tasks while visualizing the relationships between those tasks is fundamental to the process that allows systems engineering to work. The challenge is to decide which things to simplify, but often these are self-evident upon inspection of the project. Abstractions of the project can also aid in tasks such as prioritization, where a designer may look at the task and see what the trade-offs would be for any given change or development. An example of this is often seen in computer programming, where data transactions are abstracted to the data structure. A data structure is the way in which the data is organized; it is the relationship between one datum and another, be it linear or an ever-expanding tree. Once the processes have been abstracted, a systems engineer can devise functions to accomplish these goals in the abstract, and then pass in the specifics.

Holistic View. A holistic view of the project is also necessary to good results in systems engineering, because every component must fit together and no part can be viewed in isolation. Considering how the end product fits together is essential to successful completion of the goals. The process by which the product is designed is itself designed and must be optimized and changed as needed. By recognizing the design process as designed, systems engineers can ensure that they are achieving optimal results and setting useful goals. Through awareness of these facts and a good view of how the whole project fits together, systems engineers can spot the best ways to interface and prioritize the project and its subprojects.

Models. When dealing with complex and often abstract systems, it is often imperative to have some type of model. Not simply useful for the engineers who use them to keep track of their place in the larger picture, models also serve as invaluable tools for communicating with the client to better ensure that the project is turning out as desired and get feedback as the work progresses.

There are many varieties of models. Graphic representations are very common and often include flow charts, which can model things ranging from outcomes of a project to the workings of a computer program to failure modes and emergency systems. Other models are simpler, such as equations, graphs, and charts. Models are useful in design as well, as they can be used for testing. Models can also be more abstract, such as the waterfall design paradigm, an approach to development. The power in these sorts of models is that they give everyone a way of organizing their tasks within a shared framework.

Often, models will have a standardized methodology, so that they can be shared across projects and participants will be familiar with the layout. One such standardization is Unified Modeling Language, or UML. Unified Modeling Language is built for use in software design. This system aids in the identification of logical groupings of components, processes, and agents in those transactions and in the development of reusable software.

System. For something so critical, it may surprise many to find that there is no single agreed-upon definition of a system. Fundamentally, a system is a set of elements that achieves a goal. The looseness of this definition allows it to be adapted to a specific problem set. This also allows for the system to be enlarged and shrunk as the project progresses, while retaining the understanding that the designers are part of the process as well

Complexity. An understanding of the complexity of the project is critical to successful systems engineering. This comes up in the hierarchy of the project as critical components are identified and prioritized, thus motivating the development of better systems, algorithms, and analytical tools to handle the complexity of a given system, such as a space station.

Effectiveness. Effectiveness is the rubric by which the success of the project is judged. It can include a series of objectives that must be optimized overall. Because of this, a project that accomplishes one of its objectives very well but the rest very poorly might be less effective than one that is mediocre across all objectives. Objectives are things that can be measured objectively and are benchmarks for progress and success. A very common objective is cost, which is often a deciding factor on the systems that can be implemented and motivates many tradeoffs. While ideally a system could use the best components for all of its parts, such items are often expensive. This forces the engineer to decide which segments of the project require the best available performance and which do not.

Applications Past and Present

Apollo Program. There are few better examples of systems engineering than NASA’s Apollo program. The project emerged out of a Cold War rivalry between the United States and the Soviet Union. Both powers were attempting to assert dominance through superiority in space technology. The Soviets sent the first human into space in 1961. Shortly afterward, President Kennedy declared the American intention of going to the moon before the end of the decade. Thus NASA’s problem was more than simply designing a new spaceship to get to the moon, as it may have appeared at first glance. Instead, they had to first get people into space and devise ways for them to live under those conditions for the mission’s duration; NASA also had to land astronauts safely on the moon and let them take off again, all while designing new rockets to supply power. Engineers needed to assess what the surface of the moon was like and devise new systems to support the rockets and astronauts. The entire American space program during that period became a test bed for developing the technologies and systems required to get to the moon. NASA engineers designed a process by which to develop the technology in a timely and efficiently manner by first determining what components were likely critical and then devising a way to test as many of these features as possible.

They began with the Mercury space program in the late 1950s, which simply aimed to get astronauts into space. Once this goal was achieved, engineers began collecting data on what happened to people during space travel. The next step was Project Gemini, a program intended to test the engineers’ ability to design systems such as airlocks and accommodations for multi-person crews and long-duration flights. Other tasks included docking in space. Once these systems were designed and tested, the work then shifted to the Apollo program, which developed the lunar technology. Simultaneously, automated satellites were designed and sent to find good landing sites on the moon. The successful integration of several successive and simultaneous programs over nearly a decade allowed the Americans to land two people on the moon in 1969. Since the end of the Apollo program and the reduction in American space-oriented ambition, NASA has needed to become more flexible in their approach.

Defense Industry. With the need in the defense industry to coordinate complex projects across multiple agencies and organizations to obtain optimal results, systems engineering is immensely important to ensure that work is carried out as efficiently as possible. This is not simply for large-scale physical installations, such as the radar network for the North American Aerospace Defense Command (NORAD), but also for intelligence and intelligence-distribution systems. Before the development of the paradigm, there was often conflict between agencies, which reduced effectiveness. The DOD benefited greatly from people outside the agency providing new perspectives and setting up better systems to allow those agencies to interact. Thus, systems engineering not only designs a system, for example a spy satellite, but also designs the context for said system, for instance a way for the Central Intelligence Agency (CIA), US Army, US Navy, and National Security Agency (NSA) to share data from that satellite. Systems engineers would also maintain this satellite in situations where, for instance, the Navy is willing to support ground stations but not fund repair missions. Thus, systems engineering designs a product and the use for that product. It is widely used throughout the military-industrial complex.

A good example of systems engineering that occurred before formalization of the field was the Manhattan Project. In order to develop nuclear weapons before Germany during World War II, the United States initiated a massive covert engineering project called the Manhattan Project. In order to make an atomic bomb, researchers first had to develop the materials for it. This required the establishment and operation of uranium-production facilities and particle accelerators for testing, as well as the procurement of facilities to test and build the bomb itself. Equally important was the need to recruit and house physicists, engineers, and laborers—even the custodial staff—in total secrecy. The need for clandestine operations caused the project to be split into many components, a good exercise in abstraction and complexity management. To ensure the success of the project in the shortest possible time, several promising approaches were used, in the hope that at least one would succeed. In support of the Manhattan Project, as part of the greater atomic war effort, missions were organized to gather data on and sabotage the German bomb project. The resulting seamless integration of science, industry, intelligence service, and international military operations is a good example of systems engineering.

Industry. The systems-engineering paradigm has spread from the military, government, and Bell Laboratories to all parts of industry. While industry does not have multiple agencies with overlapping goals to sort out, many contemporary projects are multifaceted and combine many disciplines. A good example might be the manufacture of a car. Not only must the structure of the car be designed, but the frame and the engine must work together to create an acceptable gas mileage. If the frame is too heavy, then the gas mileage will be poor, but if the frame is not strong enough, the car will be unsafe. If a certain engine is used, the car will be too expensive, but if that engine is not used, it will not be efficient. Such considerations go into the decision-making process as tradeoffs. A successful and effective product will balance all of these pressures adequately.

Equally, when designing a car, a manufacturer may choose to use certain parts because they are already in production and will make procurement and maintenance easier. A systems engineer may optimize the assembly-line procedure for a new part or alter the supply chain to simplify matters. In the twenty-first century, computer technology is near universal, so computerized control systems are also often put into cars. A systems engineer will need to get an idea of how the software fits in and what it should do. Other considerations may include where to construct the car, whether it is simpler to ship it from Japan or if it is more efficient to build a new factory in America to produce the cars. Specialists in specific fields will provide the data for many of these tasks and decisions. By coordinating all of these activities, a systems engineer can deliver an optimal product.

Software Design. Software projects often use ideas from systems engineering because most programming projects have a plethora of possible solutions. These solutions are often worked on in segments by different people, yet they must seamlessly integrate internally and be compatible with a variety of external systems. This integration is typically accomplished by a process called encapsulation, wherein each piece of the program functions as its own unit, with specified inputs and outputs. The code thus functions as a black box, where data is placed inside and the proper outputs are received without having to look into the box at all. By stacking these boxes together and sharing the list of specifications, engineers know that a piece of code, written by one person for one project, will work with code written for a wholly different project.

In fact, the object-oriented design paradigm of computer-programming languages is structured around systems-engineering ideals. In a good program, the general flow can easily be followed and the data is stored in a manner that is readily accessible when it is needed. Data is kept in discrete sections where it can be called upon by central and universal processes; the particulars are typically handled by the code surrounding those specific sections. For example, a database program would have sorting functions that work on all things that could be put in the database, but the specific types of things—whether they are, dogs, cats, or robots—would have functions specific to them. These might be things such as determining breed or seeing if it a robot has been charged today. Many object-oriented languages have data constructs such as interfaces, which list out a set of functions that are implemented by the code. In general, the systemic approach allows code to be easy to work with and easy to maintain.

Social Context and Future Development

The paradigm of systems engineering has dramatically changed a variety of fields and is the new model of integration for large-scale projects. In a world that is increasingly interdisciplinary, the ability to integrate information and make clear goals across a project is critical. With the rise of ubiquitous computing, even the most prosaic projects make use of coding; new materials give new options for formerly simple tasks. These and many other changes to the way we work mean that unification of information will be invaluable.

Systems engineering has already overseen some of the largest projects in the twentieth and twenty-first centuries, including the Apollo program and the optimization of work flow in many industries. Because programming is actually abstract data manipulation, it is a good example of systems-engineering principles. Since the design of a program is usually arbitrary—a problem can often be solved in multiple, equally efficient ways—constraints from the client and adaptability typically dominate. Projects often involve many programmers, each working on a different section. The qualities that make for good code on the individual scale are used to coordinate mass effort. This is because even the simplest projects tend to involve multiple pieces to keep them simple at the abstract level. A programmer might put code for data manipulation in one place and the interface in another. This, of course, necessitates a means for sharing, which is accomplished though variables that can then be used by other people to add to the project. This use of systemic thinking is easily applicable to any type of project. A recent trend has been the rise of consulting firms that provide systemic analysis for their clients. As humans engage in increasingly ambitious projects, designing not only the technology but also the process around the technology will lead to better results.

Bibliography

Blanchard, Benjamin S., and W. J. Fabrycky. Systems Engineering and Analysis. Boston: Prentice, 2011. Print. Focuses on the design and engineering of human-made systems and systems analysis. Concentrates on the use of systems engineering to identify what an entity is to do before determining what the entities are.

Buede, Dennis M. The Engineering Design of Systems: Models and Methods. Hoboken: Wiley, 2009. Print. Provides a reference to the methods for systems engineering that are in use today. Uses a model-based approach to introduce methods and models used in actual practice.

Eisner, Howard. Thinking: A Guide to Systems Engineering Problem-Solving. CRC, 2019.

Kerzner, Harold. Project Management: A Systems Approach to Planning, Scheduling, and Controlling. Hoboken: Wiley, 2009. Print. Often referred to as “the Bible” of project management. Aligns with the latest release of the Project Management Institute’s Project Management Body of Knowledge.

Moiz, Abdul. "What Is Systems Engineering? (With Steps and Skills.)" Indeed, 29 July 2023, www.indeed.com/career-advice/finding-a-job/what-is-systems-engineering. Accessed 25 Sept. 2023.

Rouse, William B., and Andrew P. Sage. Handbook of Systems Engineering and Management. Hoboken: Wiley, 2009. Print. Focuses on the engineering and management techniques for systems design, with emphasis on information technology and software-intensive systems involving human and organizational elements.

"Systems Engineering Handbook." NASA, www.nasa.gov/seh/index.html. Accessed 25 Sept. 2023.

Zhang, Mingjun, and Ning Xi. Nanomedicine: A Systems Engineering Approach. Singapore: Pan Stanford, 2009. Print. Provides a detailed look at nanomedicine from a systems-engineering perspective.

About the Author

Gina Hagler writes about science and technology. Her book about applied fluid dynamics, Modeling Ships and Space Craft: The Science and Art of Mastering the Oceans and Sky, was published in 2012.