Stands for generally accepted accounting
principles—a collection of rules, procedures and conventions,
established by various government and private organizations, to ensure that
a company's financial statements accurately convey its financial status.
Some vendors of project portfolio
management tools advertise that they incorporate metrics (or have other
features) that facilitate compliance with GAAP.
A branch of applied mathematics that analyzes the
behavior of individuals who are pursuing self interest against other
individuals who are doing the same. Game theory addresses situations in
which an individual or organization chooses actions in situations where the
consequences of those actions depend on the choices made by others.
Applications of game theory typically attempt to find an optimal strategy
for one player to use when an opponent is also assumed to be playing
optimally. Such strategies are termed equilibrium strategies in that they
are stable and unlikely to change. Game theory is used to support decision
making in many different areas, including business and military strategy.
Some project portfolio management tool intended for application to
investments in highly competitive environments utilize consequence models based on game
In project prioritization, the phenomenon wherein
individuals deliberately bias estimates in order to improve the evaluation
of favored projects.
A horizontal bar chart that displays the timing,
duration, and interactions among multiple, time-phased activities, tasks,
or projects. Gantt Charts, named after
the originator Henry L Gantt, are a useful project management and planning
A simple Gantt chart
As shown in the example, a Gantt chart is a essentially a
table with each row corresponding to an activity. Time (e.g., measured in
days, weeks, or months) is denoted by the columns. Each task is represented
by a bar extending across the time columns, indicating the planned duration
of the task. Milestones and critical path lines are used to add further
detail to the chart. Milestones are important checkpoints or deadlines and
are indicated by small symbols in the time columns. Critical path lines connect task bars to
indicate dependencies, such as the requirement that a task be initiated
after the completion or commencement of another task.
Gantt charting capability is routinely provided in tools
for project management and in many project portfolio management tools.
A systematic comparison of the current situation to the
desired state. Gap analysis generally includes benchmarking and other assessments for
clarifying expectations. The end goal of gap analysis is the development of
specific plan for closing the gap and moving organizational performance to
the desired state.
A mathematical search technique for solving optimization problems based on an algorithm consisting of steps similar to
those that occur in natural evolution. The technique has found applications
in numerous fields, including biology, engineering, economics, and
Basically, a genetic algorithm seeks an optimal solution
by simulating an evolutionary process. An initial "population" of potential
solutions is randomly generated. Each candidate solution in this initial
"generation" is evaluated to determine its "fitness." Those determined most
fit "survive" and are combined and mutated to form a new population (the
next generation). The new population is then similarly used to conduct the
next iteration of the algorithm. The process continues until either a
maximum number of generations has been produced or an acceptable fitness
level has been obtained for a solution.
Flowchart of the basic genetic algorithm
Optimization based on generic algorithms is applicable to
project portfolio management (ppm), and
some ppm tools use the technique. In this context, an initial generation of
potential portfolios is randomly generated (parent portfolios). Portfolios
that don't meet the constraints (e.g., cost constraints) are eliminated
(they die off). Pairs of individual portfolios are then combined (e.g., by
choosing every other project from each) to produce second-generation
portfolios (child portfolios). Child portfolios that don't meet the
constraints are eliminated. The remaining portfolios are then evaluated,
and the highest ranked are selected to represent the next generation of
portfolios. This process is repeated for a set number of iterations or
until the user-specified optimization parameters are satisfied.
A common problem with optimization algorithms is
premature convergence—the optimizer homes in on a solution that is
not really optimal. This occurs with genetic algorithms because the
population of potential solutions being used loses diversity. To address
this problem, an approach is used that mimics the way nature maintains
diversity. New portfolios ("genetic mutations") are randomly generated and
periodically introduced into the portfolio populations. The mutated
portfolios only "survive" if they meet the constraints, in which case they
help keep the genetic algorithm from converging to a false optimum.
A desired result or end to which effort is directed. The
words objectives and goals are sometimes mistakenly used used interchangeably. However, in
technical usage the word objective
typically refers to a precise, well-defined desire (expressed so as to
indicate an object of value, context, and the direction of preference),
whereas the word goal may refer to a longer-term, less concrete aim.
Objectives are measurable, whereas goals may not be. For example, learning
more about Chinese history might be a goal whereas obtaining a high score
on the Chinese history test might be an objective. In other contexts, the
word goal is used to specify a target level of achievement against an
objective, for example, "obtain a score of at least 95 on the history
An individual or organization will typically have
multiple goals. Goal factoring is the
process of structuring goals, which can aid the identification, evaluation,
and prioritization of possible
actions for achieving those goals. One first lists goals, then organizes
them in relation to each other, identifying the goals that are most desired
and those that might be complementary or conflicting. Typically, goals are
arranged graphically in a diagram with lines or arrows indicating
Example output of goal factoring
Goal factoring is basically a less-rigorous version of
the methods for modeling decision-maker preferences used by more formal
decision-aiding methodologies (such as multi-attribute utility analysis and AHP). For example, goal factoring diagrams can
appear similar to objectives
hierarchies, although the goals appearing in the diagrams typically do
not meet all of the requirements (such as being measurable,
non-overlapping, etc.) necessary to allow the diagrams to be converted to
quantitative decision models.
Even so, in some contexts (e.g., financial investment portfolios), the term
goal factoring has been used to refer to a process of deriving a single
metric that accounts for portfolio
risk, timing, return, and other relevant measures of portfolio
goal programming (GP)
An optimization method designed for problems with more
than one objective and discrete or continuous decision variables that
allows a solution to be found using linear
programming. The method, sometimes applied to resource allocation and
project selection problems, involves expressing goals or targets for
objectives and assigning priorities
or weights to achieving those targets. For example, the formulation might
be to find the subset of R&D project opportunities that comes closest
to achieving specified goals for manpower utilization, market share, sales,
and net present value maximization.
Constraints on acceptable solutions may also be defined, for example,
requiring total costs to be no greater than the budget. Goal programming
(GP) seeks solutions that meet constraints while minimizing the weighted
sum of the deviations from the specified targets. The solution effectively
involves a repetitive process of attempting to achieve each goal, in order
of priority, subject to the specified constraints.
GP has several attractive features. One obviously, is the
ability to address multiple objectives. Also, and importantly, solutions
can be easily found using the Simplex Method of linear programming.
This means that relatively large numbers of decision variables,
constraints, and goals may be established without creating difficulties for
finding a solution. For example, in the context of resource leveling, you could solve
for multi-year solutions that closely achieve many detailed goals with many
A feature that no doubt promotes the use of GP is the
(apparent) non-demanding nature of the necessary inputs. Like other
prioritization logics, a model relating choices to performance is required.
However, GP does not require detailed quantification of decision making
preferences and willingness to make tradeoffs the way that most multi-criteria analysis methods do. Instead, all
that is required is the specification of goal targets and weights.
The main disadvantage of GP, of course, is that
reasonable goals and targets cannot be specified without reference to
underlying decision-maker preferences—choosing the "right" targets
and weights is exceedingly difficult. If the targets and weights are not
appropriate, the solution will not be the one in the best interest of the
organization. The reason for this is that GP programming does not allow
tradeoffs between goals. For example, if sales growth is the first priority
goal, and market share is the second, the formulation implies that not even
one dollar of sales growth can be sacrificed to obtain even a huge gain in
market share. GP represents a "satisficing" approach to decision making,
meaning that what is sought is a satisfactory solution rather than one that
is truly optimal. Because the method does not capture willingness to
tradeoff achievement of the various objectives, it is incapable of finding
solutions that lie on the efficient frontier.
Goal programming has seen applications in production
planning, scheduling, health care, portfolio selection, distribution system
design, energy planning, water reservoir management, timber harvest
scheduling, and wildlife management problems. Many of these applications
have been used in combination with other methods to accommodate the proper
weighting of criteria.
The means by which an organization regulates and controls
organizational behavior in accordance with its goals and objectives. A
governance structure establishes accountability by implementing systems to
monitor and record what people do, includes steps for ensuring compliance
with policies, and provides for corrective action in cases where rules have
not been appropriately followed.
A decision-aiding technique that involves creating a
table with options listed as rows and factors that need to be considered as
columns. Each option/factor combination is then scored, the scores are
weighted, and the results added to provide an overall score for the option.
The approach, although simplistic, is used by some project portfolio management tools to rank
A dynamic of groups that promotes faulty decision making.
Group think, a term coined by psychologist Irving Janis in the 1970s, has
been widely studied. Consequences of group think include the tendency of
groups to overlook alternatives, selectively collect information, fail to
anticipate adverse consequences of choices, assume agreement among members
when it does not exist, and fail to develop contingency plans. In addition,
studies show that groups tend to possess a sense of invulnerability which
promotes risk taking, and causes members to stereotype the views of those
outside the group, self-censor their own views if they go against the group
view, and believe in their inherent moral superiority over those outside
Pronounced GOO-ee, GUI stands for graphic user interface
and refers to the common method used for enabling humans to interact with a
computer program. A GUI is graphic-based and obtains user inputs at least
in part via icons, pictures, and menus (which the user designates with a
mouse or other pointing device) as well as through keyboard-entered