unit 2s.pdf

7/26/2019 Unit 2s.pdf

1/36

UNIT II

MODELS OF REPRESENTING SYSTEMS:

A mathematical modelis a description of a system using mathematical concepts and

language. The process of developing a mathematical model is termed mathematical modeling.

Mathematical models are used not only in the natural sciences (such as physics, biology, earthscience, meteorology) and engineering disciplines (e.g. computer science, artificial intelligence),

but also in the social sciences (such as economics, psychology, sociology and political science);

physicists, engineers, statisticians, operations research analysts and economists use mathematical

models most extensively. A model may help to explain a system and to study the effects of

different components, and to make predictions about behavior.

Mathematical models can take many forms, including but not limited to dynamical systems,

statistical models, differential euations, or game theoretic models. These and other types of

models can overlap, !ith a given model involving a variety of abstract structures. "n general,

mathematical models may include logical models, as far as logic is taken as a part ofmathematics. "n many cases, the uality of a scientific field depends on ho! !ell the

mathematical models developed on the theoretical side agree !ith results of repeatable

experiments. #ack of agreement bet!een theoretical mathematical models and experimental

measurements often leads to important advances as better theories are developed.

Example o! mathematical model

Many everyday activities carried out !ithout a thought are uses of mathematical models.

A geographical map pro$ection of a region of the earth onto a small, plane surface is a

model !hich can be used for many purposes such as planning travel.

Another simple activity is predicting the position of a vehicle from its initial position,

direction and speed of travel, using the euation that distance travelled is the product of

time and speed. This is kno!n as dead reckoning !hen used more formally.

Mathematical modeling in this !ay does not necessarily reuire formal mathematics;

animals have been sho!n to use dead reckoning.

%opulation &ro!th. A simple model of population gro!th is the Malthusian gro!th

model. A slightly more realistic and largely used population gro!th model is the logistic

function, and its extensions.

Model of a particle in a potential-field""n this model !e consider a particle as being apoint of mass !hich describes a tra$ectory in space !hich is modeled by a function giving

its coordinates in space as a function of time. The potential field is given by a function V'

R Rand the tra$ectory is a solution of the differential euation

UNIT 2 REPRESENTATION AND ANALYSIS OF SYSTEM STRUCTURE Page 1
http://en.wikipedia.org/wiki/Dead_reckoninghttp://en.wikipedia.org/wiki/Populationhttp://en.wikipedia.org/wiki/Logistic_functionhttp://en.wikipedia.org/wiki/Logistic_functionhttp://en.wikipedia.org/wiki/Differential_equationhttp://en.wikipedia.org/wiki/Populationhttp://en.wikipedia.org/wiki/Logistic_functionhttp://en.wikipedia.org/wiki/Logistic_functionhttp://en.wikipedia.org/wiki/Differential_equationhttp://en.wikipedia.org/wiki/Dead_reckoning

7/26/2019 Unit 2s.pdf

2/36

*ote this model assumes the particle is a point mass, !hich is certainly kno!n to be false

in many cases in !hich !e use this model; for example, as a model of planetary motion.

Model of rational behavior for a consumer""n this model !e assume a consumer faces a

choice of n commodities labeled +,,...,n each !ith a market price p+, p,..., pn. The

consumer is assumed to have a cardinalutility function U(cardinal in the sense that itassigns numerical values to utilities), depending on the amounts of commoditiesx+,x,...,

xnconsumed. The model further assumes that the consumer has a budget M!hich is used

to purchase a vectorx+,x,...,xnin such a !ay as to maximi-e U(x+,x,...,xn). The problem

of rational behavior in this model then becomes an optimi-ationproblem, that is'

sub$ect to'

This model has been used in general euilibrium theory, particularly to sho! existence

and %areto efficiencyof economic euilibriums. o!ever, the fact that this particular

formulation assigns numerical valuesto levels of satisfaction is the source of criticism

(and even ridicule). o!ever, it is not an essential ingredient of the theory and again this

is an ideali-ation.

Neighbour-sensing model explains the mushroom formation from the initially chaotic

fungalnet!ork.

Computer Science' models in /omputer *et!orks, data models, surface model.

Mechanics' movement of rocket model.

Some application

A mathematical model usually describes a system by a set of variables and a set of euations that

establish relationships bet!een the variables. 0ariables may be of many types; real or integer

numbers, booleanvalues or strings, for example. The variables represent some properties of the

system, for example, measured system outputs often in the form of signals, timing data, counters,

and event occurrence (yes1no). The actual model is the set of functions that describe the relationsbet!een the different variables.

#$ilding %loc&

http://en.wikipedia.org/wiki/Optimization_(mathematics)http://en.wikipedia.org/wiki/Pareto_efficiencyhttp://en.wikipedia.org/wiki/Neighbour-sensing_modelhttp://en.wikipedia.org/wiki/Mushroomhttp://en.wikipedia.org/wiki/Mushroomhttp://en.wikipedia.org/wiki/Fungushttp://en.wikipedia.org/wiki/Computer_Sciencehttp://en.wikipedia.org/wiki/Mechanicshttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Boolean_data_typehttp://en.wikipedia.org/wiki/Signal_(electronics)http://en.wikipedia.org/wiki/Optimization_(mathematics)http://en.wikipedia.org/wiki/Pareto_efficiencyhttp://en.wikipedia.org/wiki/Neighbour-sensing_modelhttp://en.wikipedia.org/wiki/Mushroomhttp://en.wikipedia.org/wiki/Fungushttp://en.wikipedia.org/wiki/Computer_Sciencehttp://en.wikipedia.org/wiki/Mechanicshttp://en.wikipedia.org/wiki/Real_numberhttp://en.wikipedia.org/wiki/Integerhttp://en.wikipedia.org/wiki/Boolean_data_typehttp://en.wikipedia.org/wiki/Signal_(electronics)

7/26/2019 Unit 2s.pdf

3/36

There are six basic groups of variables namely' decision variables, input variables, state

variables, exogenous variables, random variables, and output variables. 2ince there can be many

variables of each type, the variables are generally represented by vectors.

3ecision variables are sometimes kno!n as independent variables. 4xogenous variables are

sometimes kno!n as parameters or constants. The variables are not independent of each other asthe state variables are dependent on the decision, input, random, and exogenous variables.

5urthermore, the output variables are dependent on the state of the system (represented by the

state variables).

6b$ectives and constraints of the system and its users can be represented as functions of the

output variables or state variables. The ob$ective functions !ill depend on the perspective of the

model7s user. 3epending on the context, an ob$ective function is also kno!n as an index of

performance, as it is some measure of interest to the user. Although there is no limit to the

number of ob$ective functions and constraints a model can have, using or optimi-ing the model

becomes more involved (computationally) as the number increases.

'lai!(ing mathematical model

Many mathematical models can be classified in some of the follo!ing !ays'

+. Linea) *" nonlinea):Mathematical models are usually composed by variables,!hich

are abstractions of uantities of interest in the described systems, and operatorsthat act

on these variables, !hich can be algebraic operators, functions, differential operators, etc.

"f all the operators in a mathematical model exhibit linearity,the resulting mathematical

model is defined as linear. A model is considered to be nonlinear other!ise.

The uestion of linearity and nonlinearity is dependent on context, and linear models may

have nonlinear expressions in them. 5or example, in a statistical linear model, it is

assumed that a relationship is linear in the parameters, but it may be nonlinear in the

predictor variables. 2imilarly, a differential euation is said to be linear if it can be

!ritten !ith linear differential operators, but it can still have nonlinear expressions in it.

"n a mathematical programmingmodel, if the ob$ective functions and constraints are

represented entirely by linear euations,then the model is regarded as a linear model. "f

one or more of the ob$ective functions or constraints are represented !ith a nonlinear

euation, then the model is kno!n as a nonlinear model.

*onlinearity, even in fairly simple systems, is often associated !ith phenomena such as

chaos and irreversibility.Although there are exceptions, nonlinear systems and modelstend to be more difficult to study than linear ones. A common approach to nonlinear

problems is lineari-ation, but this can be problematic if one is trying to study aspects such

as irreversibility, !hich are strongly tied to nonlinearity.

. Dete)minitic *" p)o%a%ilitic +tochatic,: A deterministic model is one in !hich

every set of variable states is uniuely determined by parameters in the model and by sets

http://en.wikipedia.org/wiki/Variable_(mathematics)http://en.wikipedia.org/wiki/Variable_(mathematics)http://en.wikipedia.org/wiki/Operator_(mathematics)http://en.wikipedia.org/wiki/Linearhttp://en.wikipedia.org/wiki/Linearhttp://en.wikipedia.org/wiki/Linear_modelhttp://en.wikipedia.org/wiki/Differential_operatorhttp://en.wikipedia.org/wiki/Optimization_(mathematics)http://en.wikipedia.org/wiki/Linear_equationhttp://en.wikipedia.org/wiki/Linear_equationhttp://en.wikipedia.org/wiki/Nonlinearityhttp://en.wikipedia.org/wiki/Chaos_theoryhttp://en.wikipedia.org/wiki/Chaos_theoryhttp://en.wikipedia.org/wiki/Irreversibilityhttp://en.wikipedia.org/wiki/Irreversibilityhttp://en.wikipedia.org/wiki/Linearizationhttp://en.wikipedia.org/wiki/Deterministic_systemhttp://en.wikipedia.org/wiki/Variable_(mathematics)http://en.wikipedia.org/wiki/Operator_(mathematics)http://en.wikipedia.org/wiki/Linearhttp://en.wikipedia.org/wiki/Linear_modelhttp://en.wikipedia.org/wiki/Differential_operatorhttp://en.wikipedia.org/wiki/Optimization_(mathematics)http://en.wikipedia.org/wiki/Linear_equationhttp://en.wikipedia.org/wiki/Nonlinearityhttp://en.wikipedia.org/wiki/Chaos_theoryhttp://en.wikipedia.org/wiki/Irreversibilityhttp://en.wikipedia.org/wiki/Linearizationhttp://en.wikipedia.org/wiki/Deterministic_system

7/26/2019 Unit 2s.pdf

4/36

of previous states of these variables. Therefore, deterministic models perform the same

!ay for a given set of initial conditions. /onversely, in a stochasticmodel, randomness is

present, and variable states are not described by uniue values, but rather by probability

distributions.

. Static *" d(namic:A static model does not account for the element of time, !hile adynamic model does. 3ynamic models typically are represented !ith difference

euationsor differential euations.

8. Dic)ete *" 'ontin$o$:A discrete model does not take into account the function of

time and usually uses time9advance methods, !hile a /ontinuous model does.

/ontinuous models typically are represented !ith f(t) and the changes are reflected over

continuous time intervals.

:. Ded$cti*e- ind$cti*e- o) !loating:A deductive model is a logical structure based on a

theory. An inductive model arises from empirical findings and generali-ation from them.

The floating model rests on neither theory nor observation, but is merely the invocationof expected structure. Application of mathematics in social sciences outside of economics

has been critici-ed for unfounded models. 8 ?x, y@, > ?, 8, B@, /artesian product of these sets (A x ) is a set that

consists of ordered pairs !here first element of the ordered pair belongs to set A !hereas second

element belongs to set , as sho!n belo!'

A C > ?(x,), (x,8), (x,B), (y,), (y,8), (y,B)@

A relation is some subset of this /artesian product, Applying the same concept in a real !orld

scenario, consider t!o sets *ame and Age having the elements'

*ame > ?Ali, 2ana, Ahmed, 2ara@

Age > ?+:, +B, +D, +E... :@

*o! consider a subset /#A22 of the /artesian product

/#A22 > ?(Ali, +E), (2ana, +D), (Ali, F), (Ahmed, +G)@

This subset /#A22 is a relation mathematically

GR.P/I'.L MODEL'

http://en.wikipedia.org/wiki/Stochastic_processhttp://en.wikipedia.org/wiki/Stochastic_processhttp://en.wikipedia.org/wiki/Difference_equationhttp://en.wikipedia.org/wiki/Difference_equationhttp://en.wikipedia.org/wiki/Mathematical_model#cite_note-3http://en.wikipedia.org/wiki/Mathematical_model#cite_note-3http://en.wikipedia.org/wiki/Catastrophe_theoryhttp://en.wikipedia.org/wiki/Mathematical_model#cite_note-4http://en.wikipedia.org/wiki/Stochastic_processhttp://en.wikipedia.org/wiki/Difference_equationhttp://en.wikipedia.org/wiki/Difference_equationhttp://en.wikipedia.org/wiki/Mathematical_model#cite_note-3http://en.wikipedia.org/wiki/Catastrophe_theoryhttp://en.wikipedia.org/wiki/Mathematical_model#cite_note-4

7/26/2019 Unit 2s.pdf

5/36

%robabilistic graphical models are graphs in !hich nodes represent random variables, and the

(lack of) arcs represent conditional independence assumptions. ence they provide a compact

representation of $oint probability distributions. Undirectedgraphical models, also called Markov

=andom 5ields (M=5s) or Markov net!orks, have a simple definition of independence' t!o (sets

of) nodes A and are conditionally independent given a third set, /, if all paths bet!een the

nodes in A and are separated by a node in /. y contrast, directedgraphical models also called

ayesian *et!orks or elief *et!orks (*s), have a more complicated notion of independence,

!hich takes into account the directionality of the arcs, as !e explain belo!.

Hndirected graphical models are more popular !ith the physics and vision communities, and

directed models are more popular !ith the A" and statistics communities. ("t is possible to have a

model !ith both directed and undirected arcs, !hich is called a chain graph.) 5or a careful study

of the relationship bet!een directed and undirected graphical models, see the books by %earlEE,

IhittakerGF, and #aurit-enGB.

Although directed models have a more complicated notion of independence than undirected

models, they do have several advantages. The most important is that one can regard an arc from

A to as indicating that A JJcauses77 . This can be used as a guide to construct the graph

structure. "n addition, directed models can encode deterministic relationships, and are easier to

learn (fit to data). "n the rest of this tutorial, !e !ill only discuss directed graphical models, i.e.,

ayesian net!orks.

"n addition to the graph structure, it is necessary to specify the parameters of the model. 5or a

directed model, !e must specify the /onditional %robability 3istribution (/%3) at each node. "f

the variables are discrete, this can be represented as a table (/%T), !hich lists the probability that

the child node takes on each of its different values for each combination of values of its parents.

/onsider the follo!ing example, in !hich all nodes are binary, i.e., have t!o possible values,

!hich !e !ill denote by T (true) and 5 (false).


7/26/2019 Unit 2s.pdf

6/36

Ie see that the event Kgrass is !etK (I>true) has t!o possible causes' either the !ater sprinker

is on (2>true) or it is raining (=>true). The strength of this relationship is sho!n in the table. 5or

example, !e see that %r(I>true L 2>true, =>false) > F.G (second ro!), and hence, %r(I>false L

2>true, =>false) > + 9 F.G > F.+, since each ro! must sum to one. 2ince the / node has no

parents, its /%T specifies the prior probability that it is cloudy (in this case, F.:). (Think of / asrepresenting the season' if it is a cloudy season, it is less likely that the sprinkler is on and more

likely that the rain is on.)

The simplest conditional independence relationship encoded in a ayesian net!ork can be stated

as follo!s' a node is independent of its ancestors given its parents, !here the ancestor1parent

relationship is !ith respect to some fixed topological ordering of the nodes.

y the chain rule of probability, the $oint probability of all the nodes in the graph above is

P(C, S, R, W) = P(C) * P(S|C) * P(R|C,S) * P(W|C,S,R)

y using conditional independence relationships, !e can re!rite this as

P(C, S, R, W) = P(C) * P(S|C) * P(R|C) * P(W|S,R)

!here !e !ere allo!ed to simplify the third term because = is independent of 2 given its parent

/, and the last term because I is independent of / given its parents 2 and =.

Ie can see that the conditional independence relationships allo! us to representthe $oint more

compactly. ere the savings are minimal, but in general, if !e had n binary nodes, the full $oint


7/26/2019 Unit 2s.pdf

7/36

!ould reuire 6(n) space to represent, but the factored form !ould reuire 6(n k) space to

represent, !here k is the maximum fan9in of a node. And fe!er parameters makes learning

easier.

T)ee t)$ct$)e

A t)ee t)$ct$)eis a !ay of representing the hierarchicalnature of a structurein a graphicalform. "t is named a Ktree structureK because the classic representation resembles a tree, even

though the chart is generally upside do!n compared to an actual tree, !ith the KrootK at the top

and the KleavesK at the bottom.

A tree structure is conceptual, and appears in several forms. 5or a discussion of tree structures in

specific fields, see Tree (data structure)for computer science' insofar as it relates to graph theory,

see tree (graph theory),or alsotree (set theory). 6ther related pages are listed belo!.

/lassical node9link diagrams that connects nodes together !ith line segments'

encyclopedia

/ \

science culture

/ \

art craft

De!inition

A t)eeis an undirected simple graphGthat satisfies any of the follo!ing euivalentconditions'

Gis coec!e"a" #as o c$c%es&

G#as o c$c%es' a" a si()%e c$c%e is *o+(e" i* a$ e"geis a""e" !o G&

Gis coec!e"' ,-! is o! coec!e" i* a$ sig%e e"ge is +e(o.e" *+o( G&

Gis coec!e" a" !#e 3/.e+!e0 co()%e!e g+a)#13is o! a (io+o* G&

A$ !o .e+!ices i Gca ,e coec!e" ,$ a -i-e si()%e )a!#&

"f Ghas finitely many vertices, say nof them, then the above statements are also euivalent to

any of the follo!ing conditions' Gis coec!e" a" #as 1 e"ges&

G#as o si()%e c$c%es a" #as 1 e"ges&

E0.MPLE: O)gani1ational cha)t !o) compan(

UNIT 2 REPRESENTATION AND ANALYSIS OF SYSTEM STRUCTURE Page
http://en.wikipedia.org/wiki/Hierarchyhttp://en.wikipedia.org/wiki/Structurehttp://en.wikipedia.org/wiki/Tree_structure#Representing_treeshttp://en.wikipedia.org/wiki/Treehttp://en.wikipedia.org/wiki/Tree_(data_structure)http://en.wikipedia.org/wiki/Tree_(graph_theory)http://en.wikipedia.org/wiki/Tree_(set_theory)http://en.wikipedia.org/wiki/Graph_(mathematics)#Simple_graphhttp://en.wikipedia.org/wiki/Equivalenthttp://en.wikipedia.org/wiki/Connected_graphhttp://en.wikipedia.org/wiki/Closed_walkhttp://en.wikipedia.org/wiki/Edge_(graph_theory)http://en.wikipedia.org/wiki/Complete_graphhttp://en.wikipedia.org/wiki/Minor_(graph_theory)http://en.wikipedia.org/wiki/Path_(graph_theory)http://en.wikipedia.org/wiki/Hierarchyhttp://en.wikipedia.org/wiki/Structurehttp://en.wikipedia.org/wiki/Tree_structure#Representing_treeshttp://en.wikipedia.org/wiki/Treehttp://en.wikipedia.org/wiki/Tree_(data_structure)http://en.wikipedia.org/wiki/Tree_(graph_theory)http://en.wikipedia.org/wiki/Tree_(set_theory)http://en.wikipedia.org/wiki/Graph_(mathematics)#Simple_graphhttp://en.wikipedia.org/wiki/Equivalenthttp://en.wikipedia.org/wiki/Connected_graphhttp://en.wikipedia.org/wiki/Closed_walkhttp://en.wikipedia.org/wiki/Edge_(graph_theory)http://en.wikipedia.org/wiki/Complete_graphhttp://en.wikipedia.org/wiki/Minor_(graph_theory)http://en.wikipedia.org/wiki/Path_(graph_theory)

7/26/2019 Unit 2s.pdf

8/36

Example

The example tree sho!n to the right has B vertices and B N + > : edges. The uniue simple path

connecting the vertices and B is 989:9B.

Hierarchical database model

A hie)a)chical data%ae modelis a data modelin !hich the data is organi-ed into a tree9like

structure. The structure allo!s representing information using parent1child relationships' each

parent can have many children, but each child has only one parent (also kno!n as a 23to3man(

)elationhip). All attributes of a specific record are listed under an entity type.

UNIT 2 REPRESENTATION AND ANALYSIS OF SYSTEM STRUCTURE Page
http://en.wikipedia.org/wiki/Data_modelhttp://en.wikipedia.org/wiki/Tree_data_structurehttp://en.wikipedia.org/wiki/File:Hierarchical_Model.svghttp://en.wikipedia.org/wiki/Data_modelhttp://en.wikipedia.org/wiki/Tree_data_structure

7/26/2019 Unit 2s.pdf

9/36

O)gani1ational In!o)mation Flo4"nformation flo! in an organi-ationin t!o !ays'

+. 5e)ticall(9 5lo! up and do!n among managers

4xample' %roduction supervisors constantly communicate !ith !ith production9line

!orkers and their o!n managers.

. /o)i1ontall(9 5lo! side!ays among departments

4xample' =egional sales managers from themarketingdepartment set their sales goals by

coordinating !ith productionmanagers in the production department.

O)gani1ational F$nction

Most organi-ations have departmentsthat perform five basic functions'

1& .cco$nting 3 Oeep track of all financial activities.

2& P)od$ction 3 Makes company product.

3& Ma)&eting 3 Advertises, promotes, ands sells the product.

4& /$man Reo$)ce 3 5inds and hires people and handle personnel matters.

5& Reea)ch 3 3oes product research and relates ne! discoveries to the firm7s current or

ne! products.

http://www.ankedin.co.cc/the_trowbridges/default.htmhttp://theiss.steadywebs.com/insurrection_raven/en_texts/unfinished_business.htmlhttp://xfert.co.cc/danceinfosa/disalett.htmlhttp://xfert.co.cc/danceinfosa/disalett.htmlhttp://xfert.co.cc/danceinfosa/disalett.htmlhttp://alemao.lima-city.de/collectiveact/dauve11.htmhttp://www.osina.co.cc/vaksam/nm059.htmlhttp://www.ankedin.co.cc/the_trowbridges/default.htmhttp://theiss.steadywebs.com/insurrection_raven/en_texts/unfinished_business.htmlhttp://xfert.co.cc/danceinfosa/disalett.htmlhttp://alemao.lima-city.de/collectiveact/dauve11.htmhttp://www.osina.co.cc/vaksam/nm059.html

7/26/2019 Unit 2s.pdf

10/36

Management Le*el

There are threemanagementlevels in most organi-ations'

+. S$pe)*io)

A. Manage and monitor the employees or !orkers.

. =esponsible for operational matters (day9to9day operations).

/. 4xample' production supervisor monitors materials needed to build a product.

. Middle Management

A. 3eal !ith control planning, tactical planning, and decision9making.

. "mplement long9term goals of the organi-ation.

/. 4xample' regional sales manager sets sales goals for sales in several states.

. Top Management

http://indelick.co.cc/m_valuedlets/T7M.htmlhttp://indelick.co.cc/m_valuedlets/T7M.htmlhttp://indelick.co.cc/m_valuedlets/T7M.html

7/26/2019 Unit 2s.pdf

11/36

A. /oncerned !ith long9range planning(strategic planning)

. *eed information to help them plan future gro!th and direction of the

organi-ation.

/. 4xample' vice president of marketing determines demand for current products and

sales strategies for ne! products.

In!o)mation !lo4

a& Information must flow in different directions to support the different informationneeds

of management

http://astralunion.goahost.co.cc/militaryincompetence/militarismsucks.htmhttp://astralunion.goahost.co.cc/militaryincompetence/militarismsucks.htmhttp://bthclan.avafreehost.com/xwus1/pcjan08.htmlhttp://astralunion.goahost.co.cc/militaryincompetence/militarismsucks.htmhttp://bthclan.avafreehost.com/xwus1/pcjan08.html

7/26/2019 Unit 2s.pdf

12/36

,& !ach level of management has different information needs

+. St)ategic Need o! Top3le*el manage)

A. "nformation that reveals overall condition of the business in capsule form.

. "nformation from all departments belo! and from outside the

organi-ation.

/. "nformation to plan for long9range events.

3. 4xample' planning for ne! facilities

. Tactical Need o! Middle3le*el manage)

A. 2ummari-ed information (!eekly or monthly reports).

. "nformation both hori-ontal and vertical across functional lines !ithin the

organi-ation.

/. istorical, internal information to develop budgets and evaluate

performances.

3. 4xample' developing production goals, concurring !ith top9level

managers and supervisors

. Ope)ational Need o! S$pe)*io)


7/26/2019 Unit 2s.pdf

13/36

A. 3etailed current day9to9day information.

. "nformation flo! is primarily vertical.

/. /ommunicate mainly !ith middle managers and !orkers beneath them.

3. 3ay9to9day internal information to keep operations running smoothly.

4. 4xample' monitoring current supplies, current inventory, and production

output.

St)ateg( .nal(i: P)oce Flo4

Hnderstanding process flo! !ithin your organi-ation allo!s for a greater visibility into ho!

things actually get done across different $obs and departments. "t makes clear !hich activities

have Kal!ays been doneK but aren7t really adding any real value to your customers. And it

exposes things that you should be doing to help move your organi-ation for!ard in its goals and

ob$ectives.


7/26/2019 Unit 2s.pdf

14/36

5or some organi-ations, their primary resources are ra! materials, and physical assets like

plant and euipment. They use their euipment to process ra! materials for sale to

customers. At each step of the !ay as a piece of material or part goes do!n their

assembly line, there is some !ork that is done. Maybe connecting t!o parts together,

maybe polishing, may be sorting. 2mart companies analy-e every step of the process and

ask themselves, Kis this activity adding value for the end customerK. 2ome call this Kvalue

mappingK, others call it Kvalue stream engineeringK. Ihatever the fancy term, the idea

!as pioneered by Papanese manufacturing and companies like Toyota. To better

understand process flo!, and implement process flo! improvements in your organi-ation,

there are three simple uestions you must ask'

6hat i the )eal *al$e that 4e p)o*ide o$) c$tome)7

The best organi-ations are constantly asking their customers !hat they like and 36*7Tlike about the products and services they purchase. They are asking their customers !hat

else they are !anting. And most importantly, they are thinking through their customers

business and creatively finding ne! !ays of solving problems for them. "ronically, this

means at times they decide not to service a customer if it means doing something that is

outside their core competency of !hat they can do excellently !hile still making money.

6hat acti*itie ho$ld 4e eliminate that a)en8t adding *al$e7

6nce you identify !hich products and services that are creating real value for your

members, you then !ork back!ard do!n the value chain ending at the beginning of eachprocess, analy-ing each point !here there is some sort of KprocessingK going on. 6ne tool

to accomplish this is to create a %rocess 5lo! diagram !ith s!im lines. The process flo!

describes the end value added product or service and then documents every step !ithin the

organi-ation !here someone is involved in this process. 4ach process sits in a s!im line;

each s!im line represents a person (or maybe group) that is responsible for accomplishing

that particular !ork. 2o, if you !ere analy-ing your event registration process, you !ould

document every place information flo!s through the organi-ation to the end customer99

either through paper or digital means. 6nce you have created your process flo! diagram,

you should have some pretty clear !asteful activities that you can get rid of " Fo) To(ota-

and lean man$!act$)ing companie- the( ha*e identi!ied e*en 4ate!$l acti*itie-

and the( 4o)& ha)d to eliminate them !)om an( p)oce" The To(ota p)od$ction

(tem identi!ie e*en &ind o! 4ate: O*e)p)od$ction- dela(- t)anpo)t- ext)a

p)oceing- ext)a in*ento)(- 4ated motion- and ma&ing de!ecti*e pa)t"9 In an

o!!ice en*i)onment the)e a)e *e)( imila) 4ate.Qou might have a customer !aiting on


7/26/2019 Unit 2s.pdf

15/36

the phone to register. Qou might have a paper registration moving across t!o desks, and

then into a spreadsheet. Qou might find out that people are creating confirmation letters

manually. 3on7t assume anything as you do your analysis, really get the facts of ho!

people do their $ob.

/o4 can o!t4a)e aid in adding *al$e and eliminating 4ate7

A simple example of !aste !e see over and over in associations is that of data duplicate

data entry. 2ometimes this is caused by multiple soft!are systems, other times it is

caused by a Kdouble checkingK process !here one person does the !ork and another

redoes the !ork $ust to make sure it is correct. This activity leads to t!o potential !astes.

5irst, is extra processing, because t!o people are doing the same thingR The second is

re!ork, !here you have inaccurate data that needs to be corrected because the data is

being KtouchedK by too many people. These activities provide no value to your member,and are not a something that they !ould be !illing to pay extra for in their membership.

Qou can7t add a line item to their dues that says, KAdded data entryK.

"t is important to note that good process flo! organi-ation must supersede soft!are

implementation. 2mart organi-ations reali-e that spending money on soft!are and

computers to aid a broken process only makes things Kbreak fasterK. Associations need to

learn this lesson as !ell. The number one reason for soft!are implementation failure is

feature creep, !hich is the process of soft!are trying to do too much. Qou end up !ith

soft!are that is too expensive and is too complicated. "f you have done good process

mapping in your organi-ation, chances are you have already been putting your peoplethrough changes that help them do their $ob better. Training people is much cheaper than

changing soft!are to fit !hat you think you need. And many times !hat you think you

need is not !hat you actually need. A good honest process flo! assessment ensures your

activities are really adding value to your members.


7/26/2019 Unit 2s.pdf

16/36

P)oce !lo4 mapping

Many organi-ations, both expectant and small, carry on passing on the fragments of important

policies and processes to employees through !ith one9off emails and the like. %rocess limpidity

is unmatchable of the ternion identify focuses "ndiana organi-ational design, on !ith the great

un!ashed and technology. Iith the management of random appendage plebeian in!ard many

companies, not!ithstanding it is no enuire that employees are struggling to bash amp unspoilt

$ob. atomic number an bill is processed, managed client ill operating room applied skill

draught sanctioned indium many organi-ations depends 2ir Thomas More than atomic number

does and !hat /larence 2hepard 3ay Pr. of the !eek has been made on !ell9grounded intelligent

kinda than business. Ihere it lacks the clearness of mental process and personal idiosyncrasies,and political maneuvering postulate over.

"n addition, search indicates that less than FS of !are defects and religious service problems ar

imputable to random factors, such arsenic malicious employees, breakdo!n of machinery and in

the buff materials. The early EFS operating theater 2ir Thomas More of the problems is

ascribable to systemic !eaknesses !ith processes. Thus, eventide if the mathematical function of


7/26/2019 Unit 2s.pdf

17/36

business organi-ation processes is relatively simple to serve and does non need expensive cap

expenditures, it pays robdingnagian dividends "ndiana efficiency and employee commitment. "f

you are mentation almost map your processes, here are x name pointers to continue oosier 2tate

mind.

+. regard employees !ho actually coiffure the !ork on in!ards functionemployees !ho ans!er the literal mold ar nuclear number 8G the outdo lieu to scre! the

elaborate stairs in!ards apiece process. They are besides more than familiar(p) !ith the vulgar

roadblocks and bottlenecks and describe contacts atomic number 8G your administration to find

things done. reuire the employees ahead inviting them to $uncture teams of !ork on mapping.

celebrate managers and supervisors out of doors of serve map sessions, every bit rich person amp

disposition to overlook the =oger untington 2essions !ith his o!n experience.U

. key out the s!ear out get do!n and remnant activities1

for from each one process, understandably key out the first and end. "f the team ignores this

important mistreat nuclear number E: the rootage of apiece seance mapping, the ebullience of

the team, redundant activities speedily leave creeping into the picture until the s!ear out

becomes unmanageable. conceive of an body process that triggers the process, such arsenic an

bill that appear oosier 2tate type A tray. This is the beginning. and so reckon of the live on

activity. /an, for example, poster an particular for the !orld9!ide #edger.

. key out the ob$ective cognitive process and inputs and outputs1

this is !here the operate begins to take on on antiophthalmic factor ne!ly signification for

employees. The team dra!ing card should postulate employees because it is course apiece !ork

and !hat ar the expected results of apiece process. *ot entirely does this aid to nidus care on

removing non9value adding activities, $ust besides gives employees a horse sense of aim in!ard

their functional lives.

Asking teams to describe the comment to the operation and outputs provided !ill attend to to

clear up !hat necessarily s!ear out ahead it put up initiate and !hat customers ad$acent !ork

gets in front it fire begin. 5or example, agreeing that thingama$ig gathering cannot set out until

the connection scre!s ar provided it leave baseless a unit tidy sum of lick in!ard progress.


7/26/2019 Unit 2s.pdf

18/36

Methods

/e$)itic E*al$ation

Facts:

A%so ca%%e"9:e-+is!ic Re.ie' Disco-! Usa,i%i!$ Egiee+ig' Usa,i%i!$

E.a%-a!io' Use+ I!e+*ace Is)ec!io' E0)e+! Re.ie

Li*ec$c%e s!ages9A%%

A usabilityevaluation method in !hich one or more revie!ers, preferably experts, compare a

soft!are, documentation, or hard!are product to a list of design principles (commonly referred

to as heuristics) and list !here the product does not follo! those principles.

Benefts

euristicevaluation falls !ithin the category of usability engineering methods kno!n as

3iscount Hsability 4ngineering (*ielsen, +GEG). The primary benefits of these methods are that

they are less expensive than other types of usability engineering methods and they reuire fe!er

resources (*ielsen, +GEG). The beneficiaries are the stakeholders responsible for producing theproduct V it costs less money to perform a heuristic evaluation than other forms of usability

evaluation, and this !ill reduce the cost of the pro$ect. 6f course, the users benefit from a more

usable product.

http://usabilitybok.org/glossary/#usabilityhttp://usabilitybok.org/glossary/#heuristichttp://usabilitybok.org/glossary/#usabilityhttp://usabilitybok.org/glossary/#heuristic

7/26/2019 Unit 2s.pdf

19/36

Advantages

"nexpensive relative to other evaluation methods (*ielsen W Molich, +GGF).

"ntuitive, and easy to motivate potential evaluators to use the method (*ielsen W Molich,

+GGF).

Advanced planning not reuired (*ielsen W Molich, +GGF).

4valuators do not have to have formal usability training. "n their study, *ielsen and

Molich used professional computer programmers and computer science students (*ielsen

W Molich, +GGF; *ielsen, +GG).

/an be used early in the development process (*ielsen W Molich, +GGF).

5aster turnaround time than laboratory testing (Oantner W =osenbaum, +GGD).

Disadvantages

As o+igia%%$ )+o)ose" ,$ Nie%se a" Mo%ic#' !#e e.a%-a!o+s o-%" #a.e

;o%e"ge o* -sa,i%i!$ "esig )+ici)%es' ,-! e+e o! -sa,i%i!$ e0)e+!s

& M-%!i)%e e.a%-a!o+s a+e +eco((e"e" sice a sig%e

e0)e+! is %i;e%$ !o B" o%$ a s(a%% )e+ce!age o* )+o,%e(s& T#e +es-%!s *+o(

(-%!i)%e e.a%-a!o+s (-s! ,e agg+ega!e"& &

:e-+is!ic e.a%-a!io (a$ i"e!i*$ (o+e (io+ iss-es a" *ee+ (ao+ iss-es

!#a o-%" ,e i"e!iBe" i a !#i;/a%o-" -sa,i%i!$ !es!

7/26/2019 Unit 2s.pdf

20/36

As a +-%e' !#e (e!#o" i%% o! c+ea!e ?e-+e;a (o(e!s@ i !#e "esig )+ocess

7/26/2019 Unit 2s.pdf

21/36

develop a speciali-ed list of heuristics for specific audiences, like senior citi-ens, children, or

disabled users, based on a revie! of the literature.

Procedure

1& Deci"e #ic# as)ec!s o* a )+o"-c! a" #a! !as;s $o- a! !o +e.ie& Fo+

(os! )+o"-c!s' $o- cao! +e.ie !#e e!i+e -se+ i!e+*ace so $o- ee" !ocosi"e+ #a! !$)e o* co.e+age i%% )+o.i"e !#e (os! .a%-e&

2& Deci"e #ic# #e-+is!ics i%% ,e -se"&

3& Se%ec! a !ea( o* !#+ee !o B.e e.a%-a!o+s

7/26/2019 Unit 2s.pdf

22/36

A %is! o* !as;s a"o+ !#e co()oe!s o* !#e )+o"-c! !#a! $o- a! is)ec!e"

As a +es-%!' (ao+ -sa,i%i!$ iss-es (a$ ,e o.e+%oo;e"&

E.a%-a!o+s "o o! *-%%$ -"e+s!a" !#e #e-+is!ics&

E.a%-a!o+s (a$ +e)o+! )+o,%e(s a! "ie+e! %e.e%s o* g+a-%a+i!$

7/26/2019 Unit 2s.pdf

23/36

Facts:

&lso called'Affinity mapping, affinity process, OP method (the process of affinity diagramming

is derived from the OP method)

(ifec#cle stage'=euirements

Contributors'*igel evan, Oaren 2hor, /hauncey Iilson. 6riginally based on the Hsability*et

description

Version'B1FFG

Affinity diagramming is a participatory method !here concepts !ritten on cards are sorted into

related groups and sub9groups. The original intent of affinity diagramming !as to help diagnose

complicated problems by organi-ing ualitative data to reveal themes associated !ith the

problems.

4xisting items and ne! items identified by individuals are !ritten on cards or sticky notes !hich

are sorted into categories as a !orkshop activity. Affinity diagramming can be used to'

Aa%$se B"igs *+o( Be%" s!-"ies

I"e!i*$ a" g+o-) -se+ *-c!ios as )a+! o* "esig

Aa%$se B"igs *+o( a -sa,i%i!$ e.a%-a!io

uilding an affinity diagram is a !ay to interpret customer data and'

S#o !#e +age o* a )+o,%e(

Uco.e+ si(i%a+i!$ a(og )+o,%e(s *+o( (-%!i)%e c-s!o(e+s

Gi.e ,o-"a+ies !o a )+o,%e(

I"e!i*$ a+eas *o+ *-!-+e s!-"$

Methods: Brainstorming

Facts:

&lso called'/reative Thinking, Thought 2ho!ers, #ateral Thinking

(ifec#cle stages'All

Version'+1F+F

rainstorming is an individual or group process for generating alternative ideas or solutions for a

specific topic.

&ood brainstorming focuses on the uantity and creativity of ideas' the uality of ideas is much

less important than the sheer uantity. After ideas are generated, they are often grouped into

categories and prioriti-ed for subseuent research or application. The outcomes of brainstorming

are'

http://www.usabilitynet.org/tools/affinity.htmhttp://www.usabilitynet.org/tools/affinity.htmhttp://www.usabilitynet.org/tools/affinity.htmhttp://www.usabilitynet.org/tools/affinity.htm

7/26/2019 Unit 2s.pdf

24/36

A %is! o* i"eas o+ so%-!ios +e%a!e" !o a )a+!ic-%a+ )+o,%e(&

T#e i"eas o+ so%-!ios o+gaiHe" i!o g+o-)s&

So(e *o+( o* )+io+i!iHa!io ,ase" o a!!+i,-!es %i;e cos! a" *easi,i%i!$&

Benefts% Advantages and Disadvantages

Benefts

Ma$ i"eas ca ,e gee+a!e" i a s#o+! !i(e&

Re-i+es *e (a!e+ia% +eso-+ces&

T#e +es-%!s ca ,e -se" i((e"ia!e%$ o+ )+ese+.e" *o+ )ossi,%e -se i o!#e+

)+oec!s&

Advantages

Is a "e(oc+a!ic a$ o* gee+a!ig i"eas

7/26/2019 Unit 2s.pdf

25/36

centered design process, it is commonly used !hen developing a site architecture but has also

been applied to developing !orkflo!s, menus, toolbars, and other elements of system design.

Ca+" so+!ig (a$ ,e co"-c!e" as a %o !ec# (e!#o" -sig i"e0 ca+"s o+

)os!/i! o!es' o+ (a$ ,e a-!o(a!e" -sig oe o* se.e+a% so*!a+e )ac;ages

Ca+" so+!ig (a$ ,e co"-c!e" as a se+ies o* i"i.i"-a% e0e+cises' as acoc-++e! ac!i.i!$ o* a s(a%% g+o-)' o+ as a #$,+i" a))+oac# #e+e i"i.i"-a%

ac!i.i!$ is *o%%oe" ,$ g+o-) "isc-ssio o* i"i.i"-a% "ie+eces

Ca+" so+!ig is -s-a%%$ co"-c!e" as a s)eciBc ac!i.i!$ i !#e ea+%$ "esig

)#ase o* a )+oec! *o+ "eBig a a+c#i!ec!-+e' ,-! ca si(i%a+%$ ,e -se"

"-+ig a )+o"-c! e.a%-a!io !o "e!e+(ie i* -sa,i%i!$ iss-es a+e "-e !o

)+o,%e(s i!# g+o-)ig o+ g+o-) %a,e%s

2orting and grouping have long been studied !ithin psychology and the research dates back at

least to the +G:Fs. *umerous, non9peer revie!ed descriptions, case studies, and blogs have been

!ritten in the last several years on the techniue and its use in the user9centered design process,but only a fe! peer revie!ed articles on the techniue have been published and little is kno!n of

its validity or reliability as a means of directly producing a useful and usable architecture.

"nstead, card sorts are generally used to provide insight that is used by a practitioner to generate

an architecture.

Benefts% Advantages and Disadvantages

Advantages

Si()%e Ca+" so+!s a+e eas$ *o+ !#e o+gaiHe+ a" !#e )a+!ici)a!s&

C#ea) T$)ica%%$ !#e cos! is a s!ac; o* i"e0 ca+"s' s!ic;$ o!es' a )e o+)+i!ig %a,e%s' a" so(e !i(e&

-ic; !o e0ec-!e i! is )ossi,%e !o )e+*o+( se.e+a% so+!s i a s#o+! )e+io" o*

!i(e' #ic# )+o.i"es sigiBca! a(o-! o* "a!a&

Es!a,%is#e" T#e !ec#i-e #as ,ee -se" *o+ o.e+ 18 $ea+s' ,$ (a$

"esige+s&

I.o%.es -se+s eca-se !#e i*o+(a!io s!+-c!-+e s-gges!e" ,$ a ca+" so+!

is ,ase" o +ea% -se+ i)-!

P+o.i"es a goo" *o-"a!io *o+ !#e s!+-c!-+e o* a si!e o+ )+o"-c!&

Disadvantages

Does o! cosi"e+ -se+s !as;s Ca+" so+!ig is a i#e+e!%$ co!e!/ce!+ic

!ec#i-e& I* -se" i!#o-! cosi"e+ig -se+s !as;s' i! (a$ %ea" !o a

i*o+(a!io s!+-c!-+e !#a! is o! -sa,%e #e -se+s a+e a!!e()!ig +ea% !as;s&

Res-%!s (a$ .a+$ T#e ca+" so+! (a$ )+o.i"e *ai+%$ cosis!e! +es-%!s ,e!ee

)a+!ici)a!s' o+ (a$ .a+$ i"e%$&


7/26/2019 Unit 2s.pdf

26/36

Aa%$sis ca ,e !i(e cos-(ig T#e so+!ig is -ic;' ,-! !#e aa%$sis o* !#e

"a!a ca ,e "iKc-%! a" !i(e cos-(ig' )a+!ic-%a+%$ i* !#e+e is %i!!%e

cosis!ec$ ,e!ee )a+!ici)a!s&

Ma$ ca)!-+e ?s-+*ace@ c#a+ac!e+is!ics o%$ Pa+!ici)a!s (a$ o! cosi"e+

#a! !#e co!e! is a,o-! o+ #o !#e$ o-%" -se i! !o co()%e!e a !as; a"

(a$ -s! so+! i! ,$ s-+*ace c#a+ac!e+is!ics

Appropriate Uses

/ard sorting can be used to'

I"e!i*$ !#e(es o+ )a!!e+s *+o( -a%i!a!i.e "a!a

De.e%o) !#e i*o+(a!io a" a.iga!ioa% a+c#i!ec!-+e *o+ a Je, si!e o+

a))%ica!io

Desig o+ +e"esig a si!e o+ a))%ica!io

O+gaiHe icos' i(ages' (e- i!e(s' a" o!#e+ o,ec!s i!o +e%a!e" g+o-)s

De!e+(ie #o a s)eciBc i"i.i"-a% c%assiBes i!e(s *+o( a )a+!ic-%a+ "o(ai

E0a(ie #o "ie+e! g+o-)s

7/26/2019 Unit 2s.pdf

27/36

organi-ational problem actually solved. This paper offers an operational definition of an

organi-ational decision process and relates this process to the anatomy of mathematical models

purported as being representative of it. "mplications regarding the potential of the decision

process models as tools of organi-ational design are explored.

Aggregation9 Se%ec!ig !#e "a!a i g+o-) o* +eco+"s is ca%%e" agg+ega!io& T#e+ea+e B.e agg+ega!e s$s!e( *-c!ios !#e$ a+e .iH& S-(' Mi' Ma0' A.g' Co-!& T#e$

a%% #a.e !#ei+ o )-+)ose Deco()osi!io9 Se%ec!ig a%% "a!a i!#o-!

a$ g+o-)ig a" agg+ega!e *-c!ios is ca%%e" Deco()osi!io& T#e "a!a is se%ec!e"'

as i! is )+ese! i !#e !a,%e&

&enerali'ation9 #i%e gee+a%iHa!io see(s !o ,e si()%iBca!io o* "a!a' i&e& !o

,+ig !#e "a!a *+o( U/o+(a%iHe" *o+( !o o+(a%iHe" *o+(&

Aggregation: A coce)! #ic# is -se" !o (o"e% a +e%a!ios#i) ,e!ee a

co%%ec!io o* e!i!ies a" +e%a!ios#i)s& I! is -se" #e e ee" !o e0)+ess a

+e%a!ios#i) a(og +e%a!ios#i)s&

Design pattern: aggregation

2ometimes a class type really represents a collection of individual components. Although this

pattern can be modeled by an ordinary association, its meaning becomes much clearer if !e use

the HM# notation for an agg)egation.

!xample'a small business needs to keep track of its computer systems. They !ant to record

information such as model and serial number for each system and its components.

X A very naYve !ay to do this !ould be to put all of the information into a single class type. Qou

should recogni-e that this class contains a set of repeated attributes !ith all of the problems ofthe phone bookU pattern. Qou could fix it as !e sho! here'

Inco))ect model +4ith imp)o*ement,

X The improved model !ill accommodate the addition of more types of components (a scanner,

perhaps), a system !ith more than one monitor or printer, or a replacement component on the

shelf that donZt belong to any system right no!. ut HM# allo!s us to sho! the association in a

more semantically correct !ay.


7/26/2019 Unit 2s.pdf

28/36

#ette) model +4ith UML agg)egation,

X The system is an aggregation of components. "n HM#, aggregation is sho!n by an open

diamond on the end of the association line that points to the parent (aggregated) class. There is

an implied multiplicity on this end of F..+, !ith multiplicity of the other end sho!n in the

diagram as usual. To describe this association, !e !ould say that each system is composed of

one or more components and each component is part of -ero or one system.

X 2ince the component can exist by itself in this model (!ithout being part of a system), the

system name canZt be part of its %O. Ie7ll use the only candidate key ?type, mfgr, model, 2*@ as

%O, since this class is not a parent. The system name, $ust an 5O here, !ill be filled in if thiscomponent is installed as part of a system; it !ill be null other!ise.

"n HM#, there is a stronger form of aggregation that is called compoition. The notation issimilar, using a filled9in diamond instead of an open one. "n composition, component instances

cannot exist on their o!n !ithout a parent; they are created !ith (or after) the parent and they are

deleted if the parent is deleted. The implied multiplicity on the diamondU end of the association

is therefore +..+.

INFORM.TION .R'/ITE'TURE

I*o+(a!io A+c#i!ec!-+e is a "isci)%ie a" a se! o* (e!#o"s !#a! ai( !o i"e!i*$ a"

o+gaiHe i*o+(a!io i a )-+)ose*-% a" se+.ice/o+ie!e" a$& I! is a%so a !e+( -se"

!o "esc+i,e !#e +es-%!ig "oc-(e! o+ "oc-(e!s !#a! "eBe !#e *ace!s o* a gi.e

i*o+(a!io "o(ai& T#e goa% o* I*o+(a!io A+c#i!ec!-+e is !o i()+o.e i*o+(a!ioaccess' +e%e.ac$' a" -se*-%ess !o a gi.e a-"iece' as e%% as i()+o.e !#e

)-,%is#ig e!i!$s a,i%i!$ !o (ai!ai a" "e.e%o) !#e i*o+(a!io o.e+ !i(e& I! is

)+i(a+i%$ associa!e" i!# e,si!e "esig a" i! is "i+ec!%$ +e%a!e" !o !#e *o%%oig

)+o*essioa% "isci)%ies9 Use+ i!e+*ace "esig' co!e! "e.e%o)(e!' co!e!

(aage(e!' -sa,i%i!$ egiee+ig' i!e+ac!io "esig' a" -se+ e0)e+iece "esig&


7/26/2019 Unit 2s.pdf

29/36

I! is a%so i"i+ec!%$ +e%a!e" !o "a!a,ase "esig' "oc-(e! "esig' a" ;o%e"ge

(aage(e!&

6hat i in!o)mation a)chitect$)e7

6rganising functionality and content into a structure that people are able to navigate intuitively

doesnZt happen by chance. 6rgani-ations must recognise the importance of information

architecture or else they run the risk of creating great content and functionality that no one

can ever find"

"t also discusses the relationship bet!een information architecture and usability, in the context of

real9!orld pro$ects.

(he problem: fnding is the ne) doing

/omputer systems used to be frustrating because they did very little uite badly. %eople using

systems became frustrated because they simply !erenZt capable of doing !hat they !erereuired to do.

ut technology has progressed and no! technology can do practically !hatever people !ant it to

do. 2o !hy doesnZt everyone using a computer have a large smile on their faceR

The shear !ealth of functionality and information has become the ne! problem. The challenge

facing organisations is ho! to guide people through the vast amount of information on offer, so

they can successfully find the information they !ant and thus find value in the systemR

"ntuitive navigation doesnZt happen by chance

(he cost o* *ailure

*ot only is this extremely frustrating for users, but it has serious repercussions for organisations.

Fo+ i!+ae!s i! (eas %o a"o)!io +a!es a" s!a +e.e+!ig !o -s-))o+!e"

o/%ie +eso-+ces&

Fo+ e,si!es i!# o%ie s#o))ig *aci%i!ies i! #as a sigiBca! i()ac! o

+e.e-e& Resea+c# s-gges!s !#a! a sigiBca! -(,e+ o* s#o))ig a!!e()!s

*ai% o! ,eca-se !#e -se+ #as e.a%-a!e" !#e )+o"-c!s o oe+ a" "eci"e"

agais! a )-+c#ase' ,-! ,eca-se !#e a.iga!io s$s!e( #as *ai%e" a" -se+

ca! B" !#e )+o"-c! !#e$ a+e i!e+es!e" i&

This problem is only set to get !orse as the uantity of information available through sitesincreases. Ihat can an organisation do to increase the chances that people can successfully

navigate their site and find the information they reuireR

"hat is in*ormation architecture

"nformation architecture is the term used to describe the structure of a system, i.e the !ay

information is grouped, the navigation methods and terminology used !ithin the system.


7/26/2019 Unit 2s.pdf

30/36

An effective information architecture enables people to step logically through a system confident

they are getting closer to the information they reuire.

Most people only notice information architecture !hen it is poor and stops them from finding the

information they reuire.

"nformation architecture is most commonly associated !ith !ebsites and intranets, but it can beused in the context of any information structures or computer systems.

(he evolution o* in*ormation architecture

The term information architectureU !as first coined by =ichard 2aul Iurman in +GD:. Iurman

!as trained as an architect, but became interested in the !ay information is gathered, organised

and presented to convey meaning. IurmanZs initial definition of information architecture !as

organising the patterns in data, making the complex clearU.

The term !as largely dormant until in +GGB it !as sei-ed upon by a couple of library scientists,

#ou =osenfeld and %eter Morville. They used the term to define the !ork they !ere doingstructuring large9scale !ebsites and intranets.

"n "nformation Architecture for the Iorld Iide Ieb' 3esigning #arge92cale Ieb 2ites they

define information architecture as'

1& T#e co(,ia!io o* o+gaisa!io' %a,e%%ig' a" a.iga!io sc#e(es i!#i a

i*o+(a!io s$s!e(&

2& T#e s!+-c!-+a% "esig o* a i*o+(a!io s)ace !o *aci%i!a!e !as; co()%e!io

a" i!-i!i.e access !o co!e!&

3& T#e a+! a" sciece o* s!+-c!-+ig a" c%assi*$ig e, si!es a" i!+ae!s !o

#e%) )eo)%e B" a" (aage i*o+(a!io&

4& A e(e+gig "isci)%ie a" co((-i!$ o* )+ac!ice *oc-se" o ,+igig

)+ici)%es o* "esig a" a+c#i!ec!-+e !o !#e "igi!a% %a"sca)e&

Today IurmanZs influence on information architecture is fairly minimal, but many of the

metaphors used to describe the discipline echo the !ork done by architects. 5or example,

information architecture is described as the blueprint developers and designers use to build the

system.

#ommon problems

The most common problem !ith information architectures is that they simply mimic acompanyZs organisational structure.

Although this can often appear logical and an easy solution for those involved in defining the

architecture, people using systems (even intranets) often donZt kno! or think in terms of

organisational structure !hen trying to find information.


7/26/2019 Unit 2s.pdf

31/36

Ho) to create an e+ective in*ormation architecture

An effective information architecture comes from understanding business ob$ectives and

constraints, the content, and the reuirements of the people that !ill use the site.

"nformation architecture is often described using the follo!ing diagram'

Business,#onte-t

Hnderstanding an organisationsZ business ob$ectives, politics, culture, technology, resources and

constraints is essential before considering development of the information architecture.

Techniues for understanding context include'

Rea"ig e0is!ig "oc-(e!a!io

Missio s!a!e(e!s' o+gaisa!io c#a+!s' )+e.io-s +esea+c# a" .isio

"oc-(e!s a+e a -ic; a$ o* ,-i%"ig -) a -"e+s!a"ig o* !#e co!e0! i

#ic# !#e s$s!e( (-s! o+;&

S!a;e#o%"e+ i!e+.ies

S)ea;ig !o s!a;e#o%"e+s )+o.i"es .a%-a,%e isig#! i!o ,-siess co!e0! a"

ca -ea+!# )+e.io-s%$ -;o o,ec!i.es a" iss-es&

Fo+ *-+!#e+ i*o+(a!io a,o-! s!a;e#o%"e+ i!e+.ies' see o-+ a+!ic%e

?Se%ec!ig s!a *o+ s!a;e#o%"e+ i!e+.ies@

http://www.steptwo.com.au/papers/cmb_interviewselect/index.htmlhttp://www.steptwo.com.au/papers/cmb_interviewselect/index.html

7/26/2019 Unit 2s.pdf

32/36

#ontent

The most effective method for understanding the uantity and uality of content (i.e.

functionality and information) proposed for a system is to conduct a content inventory.

/ontent inventories identify all of the proposed content for a system, !here the content currently

resides, !ho o!ns it and any existing relationships bet!een content.

/ontent inventories are also commonly used to aid the process of migrating content bet!een the

old and ne! systems.

4ffective "A must reflect the !ay people think

Users

An effective information architecture must reflect the !ay people think about the sub$ect matter.

Techniues for getting users involved in the creation of an information architecture include'

Ca+" so+!ig/ard sorting involves representative users sorting a series of cards, each labelled !ith a

piece of content or functionality, into groups that make sense to them. /ard sorting

generates ideas for ho! information could be grouped and labelled.

5or further information about card sorting, see the article /ard sorting a definitive

guideU

Ca+"/,ase" c%assiBca!io e.a%-a!io

/ard9based classification evaluation is a techniue for testing an information architecture

before it has been implemented.

The techniue involves !riting each level of an information architecture on a large card,

and developing a set of information9seeking tasks for people to perform using the

architecture.

5or further information about card9based classification evaluation, see the article

/ardbased classification evaluationU

St$les o* in*ormation architecture

There are t!o main approaches to defining an information architecture. These are'

To)/"o i*o+(a!io a+c#i!ec!-+e

This involves developing a broad understanding of the business strategies and user needs,

before defining the high level structure of site, and finally the detailed relationships

bet!een content.

o!!o(/-) i*o+(a!io a+c#i!ec!-+e

http://www.steptwo.com.au/papers/ext_cardsorting/index.htmlhttp://www.steptwo.com.au/papers/ext_cardsorting/index.htmlhttp://www.steptwo.com.au/papers/ext_cardbased/index.htmlhttp://www.steptwo.com.au/papers/ext_cardsorting/index.htmlhttp://www.steptwo.com.au/papers/ext_cardsorting/index.htmlhttp://www.steptwo.com.au/papers/ext_cardbased/index.html

7/26/2019 Unit 2s.pdf

33/36

This involves understanding the detailed relationships bet!een content, creating

!alkthroughs (or storyboards) to sho! ho! the system could support specific user

reuirements and then considering the higher level structure that !ill be reuired to

support these reuirements.

oth of these techniues are important in a pro$ect. A pro$ect that ignores top9do!n approachesmay result in !ell9organised, findable content that does not meet the needs of users or the

business. A pro$ect that ignores bottom9up approaches may result in a site that allo!s people to

find information but does not allo! them the opportunity to explore related content.

Take a structured approach to creating an effective "A

#reating an e+ective in*ormation architecture in . steps

The follo!ing steps define a process for creating an effective information architectures.

1& U"e+s!a" !#e ,-siessco!e0!-a% +e-i+e(e!s a" !#e )+o)ose" co!e!

*o+ !#e s$s!e(& Rea" a%% e0is!ig "oc-(e!a!io' i!e+.ie s!a;e#o%"e+s a"

co"-c! a co!e! i.e!o+$&

2& Co"-c! ca+"s so+!ig e0e+cises i!# a -(,e+ o* +e)+ese!a!i.e -se+s&

3& E.a%-a!e !#e o-!)-! o* !#e ca+" so+!ig e0e+cises& Loo; *o+ !+e"s i g+o-)ig

a" %a,e%%ig&

4& De.e%o) a "+a*! i*o+(a!io a+c#i!ec!-+e

7/26/2019 Unit 2s.pdf

34/36

3eveloping information architecture in this !ay enables you to design and build a system

confident that it !ill be successful.

There are many !ays to document an "A

Products *rom the in*ormation architecture process

0arious methods are used to capture and define an information architecture. 2ome of the most

common methods are'

Si!e (a)s

Ao!a!e" )age %a$o-!s

Co!e! (a!+ices

Page !e()%a!es

There are also a number other possible by9products from the process. 2uch as'

Pe+soas

P+o!o!$)es

S!o+$,oa+"s

4ach of these methods and by9products is explained in detail belo!.

Site maps

2ite maps are perhaps the most !idely kno!n and understood deliverable from the process of

defining an information architecture.

A site map is a high level diagram sho!ing the hierarchy of a system. 2ite maps reflect theinformation structure, but are not necessarily indicative of the navigation structure.

Annotated page la$outs

%age layouts define page level navigation, content types and functional elements. Annotations

are used to provide guidance for the visual designers and developers !ho !ill use the page

layouts to build the site.

%age layouts are alternatively kno!n as !ireframes, blue prints or screen details.

#ontent matri-

A content matrix lists each page in the system and identifies the content that !ill appear on that

page.


7/26/2019 Unit 2s.pdf

35/36

Page templates

%age templates may be reuired !hen defining large9scale !ebsites and intranets. %age templates

define the layout of common page elements, such as global navigation, content and local

navigation. %age templates are commonly used !hen developing content management systems.

Personas

%ersonas are a techniue for defining archetypical users of the system. %ersonas are a cheap

techniue for evaluating the information architecture !ithout conducting user research.

%rototypes can be used to bring an "A to life

Protot$pes

%rototypes are models of the system. %rototypes can be as simple as paper9based sketches, or as

complex as fully interactive systems. =esearch sho!s that paper9based prototypes are $ust as

effective for identifying issues as fully interactive systems.

%rototypes are often developed to bring the information architecture to life. Thus enabling users

and other members of the pro$ect team to comment on the architecture before the system is built.

Stor$boards

2toryboards are another techniue for bringing the information architecture to life !ithout

building it. 2toryboards are sketches sho!ing ho! a user !ould interact !ith a system to

complete a common task.

2toryboards enable other members of the pro$ect team to understand the proposed information

architecture before the system is built.

/n*ormation architecture and usabilit$

2ome people find the relationship and distinction bet!een information architecture and usability

unclear.

"nformation architecture is not the same as usability, but the t!o are closely related. As described

in a previous OM /olumn ([Ihat is usability[, *ovember FF8), usability encompasses t!o

related concepts'

Usa,i%i!$ is a a!!+i,-!e o* !#e -a%i!$ o* a s$s!e(9

!e need to create a usable intranetU

Usa,i%i!$ is a )+ocess o+ se! o* !ec#i-es -se" "-+ig a "esig a"

"e.e%o)(e! )+oec!9

!e need to include usability activities in this pro$ectU

"n both cases usability is a broader concept, !hereas information architecture is far more

specific.

http://www.steptwo.com.au/papers/kmc_whatisusability/index.htmlhttp://www.steptwo.com.au/papers/kmc_whatisusability/index.html

7/26/2019 Unit 2s.pdf

36/36

/n*ormation architecture as an attribute o* the 0ualit$ o* a

s$stem

An effective information architecture is one of a number of attributes of a usable system. 6ther

factors involving the usability of a system include'

.is-a% "esig

i!e+ac!io "esig

*-c!ioa%i!$

co!e! +i!ig&

/n*ormation architecture as a process

The process for creating an effective information architecture is a sub9set of the usability

activities involved in a pro$ect.

Although !eighted to the beginning of the pro$ect, usability activities should continuethroughout a pro$ect and evaluate issues beyond simply the information architecture.

"ho creates the in*ormation architecture1

"ncreasingly companies are realising the importance of information architecture and are

employing specialist [information architectsZ to perform this role.

ut information architecture is also defined by'

i!+ae! "esige+s a" (aage+s

e,si!e "esige+s a" (aage+s

.is-a% "esige+s

o!#e+ )eo)%e "esigig i*o+(a!io s$s!e(s

)+og+a((e+s

%i,+a+ias

!ec#ica% +i!e+s

unit 2s.pdf

Documents