[Twoey Award]
Twoey's Coolest Links
Engineer Your Love Life

A Systems Engineering Approach to Dating and Relationships

In the fall of 1995, I was taking a class in systems engineering management, and we had to write a term paper analyzing a current project we were involved in. It could be any type of project, from work related to building a house. So I decided to base my paper on my real life experiences with dating and finding my eventual spouse. I really did follow this approach!

One of my basic philosophies is yes, you do have to pay attention to your partner, but if a relationship is real "work", then you probably picked the wrong partner in the first place. You won't have this problem if you follow good systems engineering practice. Do most of your analysis up front, and make sure of your requirements before design and production. It's always easier (and CHEAPER, hint, hint - divorce) to correct the program bugs before you release the product than to keep sending out "fixes" after it's in the field.

In case you're wondering what my background/experience is in all this, well, for the past 20-25 years I've basically done two things - be an engineer and chase women. Over the years, I estimate I met anywhere from 750 to 1000 women for dates (most of these were the "one time meeting" often based on personal ads), plus get involved in about a dozen long term relationships (almost all of them very positive experiences), until I finally found Cathy. There were relationships based on being best friends, relationships based on pure physical lust (who cares if she can hold a conversation), a few lucky ones based on both, and even a couple of email/long distance relationships. So I figure I've about seen and done it all by now.

When you're finished here, I did a project for another class that dealt with Naturalistic Decision Making and Selecting a Spouse. This project applied current theories of how people make decisions to the problem of how to make a good partner choice.

Systems Engineering and Life:

Designing, Developing, and Maintaining a Permanent Relationship

December 1995


"Systems engineering is defined as the set of concepts and techniques which are necessary to proceed from the original system concept to the creation of the system or, more completely, to the satisfaction of the original need."(1) This paper will present a systems engineering management approach to an unlikely (from an engineering point of view) problem - my real life process of finding and then maintaining a primary, lifetime partnership (traditionally defined by a marriage).

In a very broad sense, this process is somewhat analogous to an evolutionary systems design and acquisition in which the most suitable commercial off-the-shelf (COTS) product (human female) must be selected to integrate into the final functioning system (marriage). COTS, as defined by the Federal Acquisition Regulation, is "an item produced and placed in stock by a contractor, or stocked by a distributor, before receiving orders or contracts for its sale."(2) On the spectrum of fully COTS to fully developmental, a person clearly falls closer to the COTS end. While you can encourage some change, for the most part, what you see is what you get.

The actual process of meeting new people and finding a partner is not necessarily a smooth one in terms of a well managed project. Seemingly good solutions can suddenly turn bad for unexpected or even unapparent reasons. There is no single project manager "in charge," and traditional configuration management of the components is virtually impossible since much of what the other person is or does is out of your control. However, there are definite parallels between the real life steps and standard systems engineering management practices, as shown in the following table:

Real Life StepSystem Engineering Equivalent
Deciding you want a permanent partner Need statement
Defining what you want in a partner and a relationshipRequirements analysis and specification
Conceptual design
Risk management plan
Determining methods for meeting potential mates Evaluation of alternatives
Cost and schedule management
Dating different candidate people Preliminary design and prototyping
Tradeoff studies
Modification of requirements/process via feedback loop
Determining you have found a possible partner Tradeoffs
Multi-criteria analysis evaluation
Detailed design, test, and evaluation
Getting engaged (or equivalent) Refinement of system operational specifications
Final acceptance testing
Getting married System deployment
Continuous operational testing
Staying employed, saving money, etc. System maintenance and support
Measures of effectiveness
Resolving differences System assessment
Modification planning and change implementation
Divorce or one person dies Planning for expected or premature system retirement

One important point needs to be made with respect to the specific relationship Cathy (my significant other) and I have forged. It is very nontraditional and unconventional as compared to society's "norm;" this is reflected in those analyses and discussions pertaining to our relationship. For instance, there was no formal "engagement," we will not wear rings or change names, and the wedding itself is simply viewed as a big party for friends and family. Many of the typical "stages" of a relationship progressing to marriage have either been blended together, reordered, or simply ignored.

Since I view this relationship so positively, there is the distinct possibility of my personal bias influencing the presentation and conclusions. I have attempted to present the subject matter as objectively as possible, either by including other people's observations and feedback, or by including measurable data if available.


In the spring of 1991, a woman I viewed as a possible permanent partner ended our relationship. During the following several months I spent a lot of time thinking about that relationship and other past failed relationships in an effort to determine what had gone wrong. I also tried to gain a better understanding of myself, and my needs. Of course, I had done a lot of these things previously, but this time I seemed a little more focused and organized about it. According to Sage(3), there are at least four management perspectives - inactive, reactive, interactive, and proactive. The basic progression from inactive to proactive deals with how quickly problems are recognized and dealt with. Previously, I had taken a more reactive and interactive approach, in which problems are dealt with either after or as soon as they arise. I made the effort to incorporate the proactive perspective, in which the attempt is made to predict the potential for trouble and to plan for its minimization.

Over the next few years of dating (with several somewhat long term, though not permanent relationships), I realized that I was at least being more successful at finding quality women (based on my requirements list outlined in the next section). Eventually I was fortunate enough to meet Cathy in the summer of 1994, and at this point (fall 1995) we are well on our way to making our relationship permanent.

Looking back on the process to date, I find that I had unknowingly been applying many of the traditional systems engineering steps along the way. Perhaps that has contributed to the success, so far, of the process.

Needs and Desires

"The system engineering process generally commences with the identification of a 'want' or 'desire' for one or more items, and is based on a real (or perceived) deficiency."(4) For me, defining the need was easy, I've always had the desire to be involved in a good relationship, and I very much wanted to find someone permanent with whom I could share a future.

There were various long term risks that could be identified. The biggest risk was making the assumption that I would eventually find someone permanent. I have been careful to not base certain life decisions on that assumption. For instance, I could have put off buying a house until it was with my eventual partner. However, financially it was advantageous to do that on my own as soon as I was able and I have done so (twice). Another risk I could identify was eventually becoming frustrated about not finding someone and then "settling" for a flawed relationship. This is a definite risk and I've seen its consequences many times in real life. To minimize it I formulated a set of requirements for the person I would like to end up with and the relationship parameters that were important to me, and planned to use this list as an evaluation tool. I also had decided long ago that given the choice of being single or being in an imperfect relationship, I would rather be single.

Picking out the proper partner is not easy. Almost half the marriages in the United States end in divorce.(5) "Are people really so dumb that they pick wrong most of the time? Probably not, what they're doing is picking on the basis of what matters to them in the short run. But what matters in the long run may be different. The factors that count change, people change, relationships change."(6) Therefore, it is important to establish the true requirements, and also a plan for managing their change.

Defining what I wanted in a partner and how I wanted us to function in a relationship was the equivalent of performing a requirements analysis and then defining a specification for both the other person and the system in which the two of us would function. Much of my analysis was based on prior dating experiences; in twenty years one can collect a lot of data. I was able to find several good books on the subject of relationships (Sills(7,8) and DeAngelis(9)) to help me organize the data and formulate my requirements. I think I gravitated to books with a well thought out, somewhat analytic approach. Indeed, many of the recommended procedures were ones that in a way I was already using.

A detailed list of all my requirements would be too lengthy to list here, but I have included the ones I feel were most important (in no particular rank order):

Partner RequirementsRelationship Requirements
Intelligent Similar Goals
AttractiveFrequent and open communication
AffectionateWilling to put time into the relationship
Good sense of humor Partners viewed and treated as equals
Similar values to my own Most decisions made mutually
Financially secure Balance of together and alone time
No children or desire for any Partners abilities complement one another
Dependable, someone I can trust

My conceptual design of the ideal partnership was one where the two people were best friends as well as being wildly passionate about each other, a partnership where both people had similar goals and lifestyles, whose strengths and weaknesses complemented each other, and who just plain "liked" being around each other. Rechtin sums up the partnership concept rather nicely in system terms: "1. A system is a complex set of dissimilar elements or parts so connected as to form an organic whole. 2. The whole is greater in some sense than the sum of the parts, that is, the system has properties beyond those of the parts. Indeed, the purpose of building systems is to gain those properties."(10)

Methods for Meeting People

There are many different ways for meeting women; each has its own distinct advantages and disadvantages. In making this analysis of alternatives, I had unknowingly used what amounted to a variation of the Pugh Concept Matrix(11) to make a decision. In this method, each alternative is assigned a value relative to a norm, and the alternative with the most positive values is chosen. If possible, positive attributes from non-chosen alternatives are then incorporated into the process. For my problem, I assigned the values low, medium, or high for each alternative depending on how well it satisfied the selection criteria.

Criteria Low cost Feeling of taking an active role High number of people to choose from Able to specify my requirements Good previous experience Efficient use of my free time
Do nothing, try and meet people in daily life high low medium low low low
Join a social or activity group medium medium medium medium low medium
Use a dating service low high low high low medium
Place a personal ad medium high medium high high high
Answer personal ads high high low medium high high
Ask friends to set me up high medium low high low high
Go to dances or clubs medium medium medium low low low

What I ended up doing in reality was to rely almost exclusively on some variation of personal ad, either locally or on the internet. The assigned values in the Pugh Concept Matrix confirm that using this approach was my preferred alternative.

There was also an issue of cost and schedule management to this process. Cost for me was defined more in time commitment rather than in a dollar amount. I had a very good idea of approximately how long I was willing to invest in each phase of the process, and set various milestones. I was willing to initially see a person 3 or 4 times to decide whether or not we even wanted to start dating semi-seriously. I expected it to become exclusive within a month, and in no more than 3 months both people should agree that there is a real possibility that things could be permanent. At 6 months I expected the decision to be made that yes, we think this will be permanent, and to start taking definite steps in that direction. My opinion is that by the time you reach your 30's, if you can't decide in 6 months, then you should stop wasting the other person's time and try someone else.

This schedule may appear to be rather rigid and inflexible. Was I prepared to make a change to it (configuration management) if necessary? The short answer was no, other than perhaps to alter the schedule by a few weeks if necessary. I take a very direct and straightforward approach to life, and this schedule was an accurate reflection of who I am, what approach works best for me, and some of the personality traits/styles of my desired partner. I was willing to take the risk of applying these schedule constraints to my search, and possibly ruling someone out prematurely.


One aspect of evolutionary systems design is that of constantly refining and modifying the requirements based on feedback from using the system in its present state of development. As Beam states, "Since the system may be substantially different in initial and later phases of evolution, the requirements placed on it in the earliest implementation may be modest and easily obtainable. As it grows in function and scope, feasibility considerations that were of no concern in the initial phase may become serious barriers to further evolution."(12) This concept is particularly relevant here, since the expectations and goals at the very beginning of a dating relationship are likely to be fewer and easier to obtain than those of a more established relationship that is progressing toward marriage.

For my process, the phase of dating different candidate people was the systems engineering equivalent of preliminary design and prototyping. This particular process presents an interesting situation not typically desirable, but impossible to avoid. Each subsequent prototype does not necessarily represent an improvement over the previous version with respect to the final requirements. In many ways dating is a trial and error process and at any given time the prototype in use may turn out to be the final design.

A major systems engineering task performed while in the dating phase were tradeoff studies to refine the system requirements. What I had to do was to determine which qualities or traits I absolutely required, and which I was willing to compromise on. As stated before, people are like COTS products, the aim is to find the best fit from the available choices. As Sills succinctly puts it, "Life is a blueplate special. You want the chicken, it comes with the peas. You want the roast beef, you get Brussels sprouts. NO SUBSTITUTIONS."(13)

I met and dated numerous women over a period of a few years, some for extended periods of time. In the process, I discovered several new things about myself and as a result had to go back and refine or add to my initial requirements. I also became more certain as to which qualities I might be willing to trade off for others (and conversely, which I wouldn't).

For example, I met one person who satisfied quite a number of my requirements and had the added benefit of a substantial inheritance at some future time. Another person was again a close fit except for a lack of time and energy commitment to the relationship. With a third person, I attempted a long distance relationship that ultimately failed; from that experience I learned that I was simply not cut out for that type of thing and limited my searching to the local area. These experiences, along with meeting many other women, helped my to prioritize my requirements and make easier the task of determining tradeoffs.

One thing I was able to do was to adhere to my schedule planning. Even though it was difficult at the time, I took the responsibility to end the two relationships that had seemed the most promising. At an earlier age I probably would have let them continue for an indefinite period, since for the most part they were good relationships. But they were unlikely to ever progress to permanence given my requirements, and I could not afford the time cost if I wanted to maximize my possibility of achieving my goals.

On to Part 2 Send email to me: larrykahn AT alum DOT mit DOT edu Back to my home page