The use of a decision matrix to prioritize SOR items in the boat design process

Discussion in 'Boat Design' started by TANSL, Nov 30, 2017.

1. Joined: Jan 2005
Posts: 1,636
Likes: 92, Points: 48, Legacy Rep: 1110
Location: Philippines

rxcompositeSenior Member

The Decision Matrix and the SOR can never be reconciled because it is an entirely different method of analysis like a dog and a cat.

First, SOR is a design goal that goes through a Design Spiral method to test the viability. It is an OPTIMIZATION process and, along the way, compromises will have to be made to achieve the specific use. Optimization is based on available graphs or existing formula that defines the graph. Several graphs that define the feasibility is used that is why the DS cannot be simplified into a simple Faze linear boundary graph.

The simplified Decision Making/Matrix is based on product existing. It assigns weight based on the narrow band of chosen product. There may be two (or more) products that meets the criteria but are not exactly the same in terms of availability or cost of operation. In Scientific Management, the more complex method of Decision Tree method is used. Decision tree is like chess, forecasting your next move based on the last arrangement.

The DM (or DT) can be used as a subset of Design Spiral but not as a whole. For example, in the DS stage, it shows that for a given Fn and displacement, it requires a 1,000 Hp engine. Sample data of 1,000 Hp available is gathered and evaluated according to cost, maintenance, service available, ect. Note that the process is used based on available products/data.

The DS can produce a design, or several feasible designs that can be evaluated. Thus, a set of data defining the passenger capacity, displacement, speed, cost to purchase, operating cost, ect. The DM can be used to choose which is appropriate for intended use but it cannot be used in the design per se.

Angélique and Barry like this.
2. Joined: Aug 2002
Posts: 12,396
Likes: 205, Points: 63, Legacy Rep: 2031
Location: Milwaukee, WI

gonzoSenior Member

They complement each other. The design spiral is an iterative process. However, the decision matrix is very useful for each iteration. For example, a vessel needs to attain 40 knots. It then needs its displacement reduced. The decision matrix will be used to decide whether the beam is decreased, the interior is simplified or the fuel tanks are made smaller.

3. Joined: Oct 2010
Posts: 6,301
Likes: 132, Points: 63, Legacy Rep: 702
Location: Australia

Mr EfficiencySenior Member

Right, well, we have one bloke who says SOR and DM are incompatible, the other says they are complementary to each other. Both can't be right.

4. Joined: Jan 2005
Posts: 1,636
Likes: 92, Points: 48, Legacy Rep: 1110
Location: Philippines

rxcompositeSenior Member

While I may agree with complementary, the methodology is not the same. Design is scientific, based on design formulas. DM is subjective, like a grading system. To quantify, the evaluator places a merit on what s/he feels on the criteria.

Only for those who understand and use it on a regular basis.

5. Joined: Oct 2010
Posts: 6,301
Likes: 132, Points: 63, Legacy Rep: 702
Location: Australia

Mr EfficiencySenior Member

It's all done by mirrors ? Seriously, things that cannot be reconciled, are not complementary.

6. Joined: Aug 2002
Posts: 12,396
Likes: 205, Points: 63, Legacy Rep: 2031
Location: Milwaukee, WI

gonzoSenior Member

Sure they are, like men and women.

fallguy and Barry like this.
7. Joined: Aug 2002
Posts: 12,396
Likes: 205, Points: 63, Legacy Rep: 2031
Location: Milwaukee, WI

gonzoSenior Member

The customer is usually the evaluator. Even if you design something for yourself, you are evaluating what goes into the design spiral and which get modified and how much on each iteration. Without criteria, there is no basis for even starting a design.

8. Joined: Jan 2011
Posts: 31
Likes: 11, Points: 8, Legacy Rep: 28
Location: Reedville, VA

BKayJunior Member

Good afternoon, I hope you don't mind me jumping in here.

I am not a NA or designer; but I have employed decision matrix for many many years as a Special Forces officer (we quite literally live and die by them). The process is very useful in choosing between 2 or more distinctly different and complex options and I'm perfectly comfortable using it to choose between multiple different boat designs. Additionally, I spent a lot of time discussing with my spouse how Northrup Grumman/Newport News Shipbuilding used it in choosing between different aspects of the CVN-X - the "Next Generation Aircraft Carrier". Since I was not "read-on" to that program, I was not given any specifics or classified information, but we did discuss at how they used the process for choosing among different options to solve some problem. As I recall, they would identify a design problem and come up with potential solutions (I can't remember if they called the solutions "Courses of Action" or "Concepts" - but they had some military sounding name to them) - to the extent some aspects of the Concepts were mutually exclusive, they would articulate the Concept, try to make each Concept the best they could. Then rather than picking the one that sounded best or voting on the solution, they spent many hours arguing over the criteria they would use to evaluate or compare the concepts, how to measure each criteria, and the relative value of that criteria.

Yep, a whole lot of arbitrary assignment of numbers. But the value of a DECMAT is that you are arguing criteria and measurements at a more micro level - one that you can get your brain around. The first rule of DECMATS is "don't let the final number drive the decision". The key is to use the final number to weed out clearly inferior Concepts - then use humans and experienced leaders to choose among the options remaining.

I've spent so many years using them in the military that I do it in my brain anytime I have a complex decision. I have not built one on paper for my "next generation cruising boat", but I've sure done it in my head. I can't give any details about CVN-X but can provide an example of an SF operation if you are interested in an example. I'm a firm believer in the value of them and felt compelled to jump in. Sorry if this is off topic since I can't provide a DECMAT for an aircraft carrier.

Dave

Angélique, TANSL and gonzo like this.
9. Joined: Jan 2011
Posts: 31
Likes: 11, Points: 8, Legacy Rep: 28
Location: Reedville, VA

BKayJunior Member

Sorry, I'm a lurker here and don't know how to link threads. But the question posed in the thread below has "DECMAT me" written all over it.

Power Catamaran vs Power Foil vs Walkaround https://www.boatdesign.net/threads/power-catamaran-vs-power-foil-vs-walkaround.59453/

Three options: Power Cat, Power Foil, Walk Around

Evaluation criteria among the three choices might be 1) acquisition cost, 2) operating efficiency, 3) survivability (protection from debris), 4) passenger comfort, 5) passenger safety, and 6) maintenance cost.

Each one of these criteria needs to be well defined (a short paragraph describing what you are measuring). This is a simple matrix and some might be weighted at 1 and others at 1.5 (I don't believe in large weightings to prevent skewed results. Don't weigh something 3 unless it is really 3 times as important.)

At it's most simple, rate each of the 3 options based on each of the 6 criteria. Or even increase your criteria to 30 if there are that many factors you want to compare. If all three options rate the same among a particular criteria then throw that criteria out - if it doesn't help solve the question the eliminate the noise.

It's an iterative process. But in the end it will help you think about what is important in a systemic manner.

Regards,

rwatson, Angélique and TANSL like this.
10. Joined: Sep 2011
Posts: 4,620
Likes: 85, Points: 58, Legacy Rep: 300
Location: Spain

TANSLSenior Member

Thank you Bkay, very interesting your comments. Still need to acces the link in your secod post and read it. Finally someone, really, has used a method to prioritize options, make decisions, but unfortunately not for the design of boats (I want to highlight design).
The armies have always used various methods of help in decision-making, many of which, if not the majority, are subsequently applied in the industry.
In World War II, General Eisenhower did not want to liberate Paris when De Gaulle asked him to do so because that liberation meant delaying the end of the war for months. Finally De Gaulle won the game, Paris was liberated and the war claimed more victims (sorry if I am very rude). The American army worked with what was then called the PERT method (if I remember correctly) to make decisions. I have used the same method many times in all kinds of jobs in shipyards.
So the methods to make decisions, are not discussed here. As you say, we all make decisions based on one system or another. Many times we think that it is decided "by common sense", but that "common sense" obeys on the one hand to an instantaneous analysis that we made of our mental database, created through successes and failures, and, on the other hand, to subjective elements difficult to concrete. But it can be said, without missing the truth, that we all use the decision matrix several times a day.
Having said all this, and I am sorry to say it once again, what some of us asked for was an example applicable to the design of a ship (again I highlight the design). Because what is interesting is not knowing how the method is, which is very simple, but how the "weights" assigned to each event or characteristic are calculated. I hope your link clarifies this aspect.
To decide the weights of the decision matrix, we must use a decision matrix that allows us to decide the correct values of the weights. Sorry for the joke.

Last edited: Dec 7, 2017 at 6:09 PM
11. Joined: Oct 2010
Posts: 6,301
Likes: 132, Points: 63, Legacy Rep: 702
Location: Australia

Mr EfficiencySenior Member

The only person I spoke to on this, apart from this forum, whose opinion I respect on technical matters, was inclined to the view that decision matrices were more applicable to choosing between existing options, rather than creating something new as an option. Is designing a boat choosing between among options, or creating something new ? It could be both, but depending on the boat. But if it was always just choosing between existing options, not much would change over time with boats.

12. Joined: Aug 2002
Posts: 12,396
Likes: 205, Points: 63, Legacy Rep: 2031
Location: Milwaukee, WI

gonzoSenior Member

It is very rare to see something truly new in a boat.

13. Joined: Jan 2005
Posts: 1,636
Likes: 92, Points: 48, Legacy Rep: 1110
Location: Philippines

rxcompositeSenior Member

You seem to be confused in your own post. Both the designer/NA and the Client/Owner evaluates the design except that they use different methods.

The SOR is a "wish list" not set in stone. The DS is a Quantitative Analysis. Anything that can be quantified or measured (how long, how wide, how heavy, how much power). S/he has at its disposal, powerful analysis tool that can do regression analysis or even software that can produce results with inputs of two or more parameters. A set of parameters can easily be generated, thus boat "A", "B", "C" ect.

DM is Qualitative Analysis because the inputs are subjective (and that includes gut feel). The DM or Pugh method is an attempt to quantify to reduce the confusion. The client assigns a weight to non quantifiable features which can impact a design. The subsets boats of similar designs A,B,C, ect are given a value based on how it meets the features. The value is multiplied by the weight and added together to arrive at a merit/grade. Note that weights and value are based on the assessment of the evaluator. The more accurate the presumption, the sounder will the decisions be.

14. Joined: Jan 2005
Posts: 1,636
Likes: 92, Points: 48, Legacy Rep: 1110
Location: Philippines

rxcompositeSenior Member

Welcome BKay. Glad to have a practitioner joining.

15. Joined: Oct 2008
Posts: 5,350
Likes: 207, Points: 63, Legacy Rep: 2488
Location: Japan