Value tree analysis

instance for Value Tree Analysis

Value tree analysis is a Multi-Criteria Decision-Making (MCDM) implement by which the decision-making attributes for each choice to come out with a preference for the decision makes are weighted. Usually, choices' attribute-specific values are aggregated into a complete method.

History

Main article: Decision analysis

The concept of utility was used by Daniel Bernoulli (1738) first in 1730s while explaining the evaluation of St Petersburg paradox, a specific uncertain gable. He explained that money was not enough to measure how much value is. For an individual, however, the worth of money was a non-linear function. This discovery lead to the emerge of utility theory which is a numerical measure that indicates how much value alternative choices have. With the development of decision analysis, utility played an important role in the explanation of economics behavior. Some utilitarian philosophers like Bentham and Mill took advantage of it as a implement to build a certain kind of ethics theory either. Nevertheless, there was no possibility of measuring one's utility function. Moreover, the theory was not so important as in practice. With the time past, the utility theory gradually based on a solid theoretical foundation. People started to use theory of games to explain the behavior of those who are rational and calm when engaging with others with conflict happening. In 1944 John von Neumann and Oskar Morgenstern's Theory of Games and Economic Behavior was published. Afterwards, it emerged since it has become of the key implement researchers and practitioners from statistics and operations research use to give a helping hand to decision makers when it was hard to make a decision. Decision analysts can be separated into two sorts of utility. The attitude of decision makers towards uncertain risk are solved by risk preference.

Process

The goal of the value tree analysis process is to offer a well-organized way to think and discuss about alternatives and support subjective judgements which are critical for correct or excellent decisions. The phases of process of the value tree analysis is shown as below:

1. Problem structuring:
   • defining the decision context
   • indentifying the objectives
   • generating and identifying decision alternatives
   • creating a hierarchical model of the objectives
   • specifying the attributes
2. Preference elicitation
3. Recommended decision
4. Sentitvity analysis

Methodology

Creating a value tree

Example of creating Value Tree Analysis
Below text

Value tree was built to be an effective and essential technique for improving and enhancing goals and values by several aspects. The tree analysis displays a visual mode to problems that used to be only available in a verbal mode. Plus separate aspects, thoughts and opinions are united to a single visual representation, which gives birth to great clarity, stimulation of creative thinking, and constructive communication.

We take the steps below to create a value tree analysis with an example to help illustrate the steps:

Step1: Initial pool

Using a free brainstorming of all the values as a beginning, by which we mean all the problems which are related to the decision: the goals and criteria, the demands, etc—all the things which have relevance to decision making. Write down what each value is on a piece of paper.

(A) It is a good idea to start this value generating process with a few very open-ended questions:

• What are issues of importance in my decision?
• What are the things that matter?
• What am I looking for?
• What do I want?
• What are my passions, intentions, joys, dreams?
• What makes me happy?
• What am I afraid of?

(B) Once you run out of ideas after this very open phase, consider the following topics to help you to come up comprehensive set of values, interests, and concerns that are relevant for your decision:

• Stakeholders

Consider who is a↵ected by the decision and what their values might be. Stakeholders may be family, friends, neighbors, society, future generations, or other species, but they can be anyone who could be substantially impacted by your decision, whether intentionally or unintentionally.

• Basic human needs. Consider whether any of the following needs might be relevant to your current decision:
  • Physiological values—e.g., health & nutrition
  • Safety values—feeling secure
  • Social values—being loved and respected
  • Self-actualization values—doing and becoming what one “is fitted for”
  • Cognitive values—craving to satisfy curiosity, to know, to explain, and to understand
  • Aesthetic values—experiencing beauty

• Intangible consequences. We are most inclined to overlook intangible consequences, such as:
  • How will you feel about yourself for having made this choice?
  • How will others think of you for having made this choice?
    • A lacking awareness of such intangible consequences can easily lead to decisions that we regret. Also, if there is a disagreement between our gut feelings about a decision and a thorough analysis, it is very often an unawareness of intangible consequences that lies at the bottom.

• Pros and cons of options you already see:
  • For each option that you can think of, ask yourself what its best and worst aspects would be. These will be values.
  • Give special consideration to costs and risks. We tend to start our planning by thinking about the positive goals we hope to achieve. It takes extra e↵ort to think about the costs and risks, but thinking about them is the first step toward avoiding them.

• Future values
  • Consider future impacts, as well as those in the present. People have a strong tendency to neglect or underweight future consequences.
  • Imagine yourself in the future, perhaps on your death bed, looking back on this decision. What would be important to you?

Step2: Clustering

Once there is a lack of ideas, clustering the ideas is an effective way—move the pieces of paper around until similar ideas are grouped together.

Step3: Labeling

Labeling each group with the higher-level value that holds it together, to make each element clearer.

• As a simplified example, let's assume some initial values we came up with are SELFACTUALIZATION, FAMILY, SAFETY, FRIENDS, and HEALTH.
• HEALTH, SAFETY, and SELF-ACTUALIZATION could be grouped together and labeled “SELF”, and FAMILY and FRIENDS could be grouped together and labeled“OTHERS”.

Step4: Moving up the tree

Seeing whether these groups can be grouped into still larger groups

• SELF and OTHERS group into OVERALL VALUE.

Step5: Moving down the tree

Also seeing if these groups can be divided into still smaller sub-groups.

• SELF-ACTUALIZATION could be divided into WORK and RECREATION.

Step6: Moving across the tree

Asking themselves is a another valid way to bring new ideas to a tree, whether any additional thoughts at that level can come out(moving across the tree).

• In addition to FAMILY and FRIENDS, we could add SOCIETY.

The figure on the right side displays the final result of the (still simplified) example. The boldfaced, italicized terms represent basic values that weren’t among the ones we initially wrote down but were brought to mind as we sought to fill out the tree.

Tool

PRIME Decisions

PRIME Decisions is a decision helping implement which use PRIME method to analyze incomplete preference information. Novel features are also offered by PRIME Decisions, which gives support to interactive decision process which includes an elicitation tour. PRIME Decisions are seen as an essential catalyst for further applied work due to its practitioners benefit from M. Köksalan et al. (eds.), Multiple Criteria Decision Making in the New Millennium © Springer-Verlag Berlin Heidelberg 2001 166 the explicit recognition of incomplete information.

Web-Hipre

Web-HIPRE, a Java applet, provides help to multiple criteria decision analysis. Moreover, a normal platform is provided for individual and group decision making. People can process the model at the same time at anytime. Plus, they can easily have access to the model.It is possible to define links to other websites. All other sorts of information like geography, media files describing the criteria or alternatives can be referred to this link, which help make a better quality of decision support significantly.

Application

Some indicators obtained by process analysis are of great help to the value tree analysis. Especially in the value decomposition of internal operation indicators, the driving indicators of a first-level process indicator are usually the secondary sub-process indicators. For instance, the new product launch cycle (in terms of R&D project to production) is actually driven by two processes: R&D and testing in the company. The standardized R&D and testing process is a key success factor for improving the speed of innovation. To this end, the two process indicators development cycle, test cycle, sample acceptance and other indicators are the vital elements which drive the new product launch cycle indicators. Therefore, combining process analysis is of great significance for the decomposition of indicator value, especially for the decomposition of internal operational indicators. The instances of the main application areas are shown as below:

Application on empirical pilot study variable selection

As value tree analysis is an approach that costs and computes little, it is one the best choices for time-sensitive variable selection in empirical pilot healthcare studies. Moreover, value tree analysis offers a well-structured and strategic process for decision-making so that pilot study and patient data constraints can be accounted for and value for study stakeholders can be maximized.

Application on Coaching

Value tree analysis help creative and critical thinking and organize the thoughts in a logical way. Moreover, when a decision has come up, value tree analysis can also be a effective way to think about one's core goals and values. Afterwards, we can actively look for decision opportunities with the analysis done before.

Softwares

DA software tools and vendors

The software tools of value tree analysis are shown in the picture below: