0 ratings0% found this document useful (0 votes) 44 views7 pagesDecision Tree
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content,
claim it here.
Available Formats
Download as PDF or read online on Scribd
an Busnes Seas FYB SEM.
i) Clee similarly the exposed profit or lst fr each ofthe
levels of manutitie 10m 215,
(9) Howmn pices shoul te manutctred otha hist expt jr
Ans. (i) €30.65 (02 pce, @ 32.88
21,” group or students ses money each yearby sling souvenir oie
stadium afer cricket match between ean 3 They con by ano
thee diferent types of somes fom a spp. The sas are
‘Sependent on which team wins the match. Conditional pay fable
nde,
Types of Sowenir
T 7 TH
© ® ©
Teamawns | 1200 | — 800 300
Temawins | 350] 700 | _ 1100
(Constr the egret table
(3) P{¢*06, whic sauvenie sould the student sock?
Dectson Theory as
sociated wit the sates of nature and he pao are shown on hese branches.
‘Thisway oeprescting a decison problem is calla adeeson tree
{et ws understand howto daw adasan te by coasidoring a simple ample:
Suppose a decision making problem is givens klow
Pay af (Profit) Table
Coane oF Aston Saisot Nate
High Demand 5) | Low Danand
Prohee hy so00| 3000
Ovisure (4) 000 = 000
Probaity 06 04
‘We sartthe decison woe with decison nade drawn av asqure Sincethere are
‘wocourssofaction,twobranches wll emerge om this ole. the ed cach
‘ofthese two branches a sate of mature node dra a a cise. From ch
‘these two circles, wo sab-branchos shoving two sates of mature S, and, wi
ener
"Thus dein tree wil ook as blow:
‘Once decison wosis drawn fora given decison problem the optimum ation can
Ams. Pry Table Regret Table
Course of | Sate of Nate =] S
Action _[Awins [wine
T | 1200 | 250 a) 2
a | 00 | 700 4, | 40
m | 300 | 1100 4, | 90] 0
Probailiy | 06 | oa Protatily | 0.6 | Oe Decision nade
BOL fe, IL are 340, 400,540; Choose with minum EOL of
DECISION TREE eosin as lows:
‘A deisin ei graphical presentation of decison proces invaving
stages. It indicates al the couse of rcton, safes of nature with tcc
robes and the conditional py. dcson problem,
A eehon dee epresetd bya square and tate of nature node
‘yale. Usually the decison nodes ac muribred sing capital etry
... frome torightand he sates fate nodosa |,23,- sequela
‘eto ight. The diferent coarsest ston rg a branches fom the dt
‘ode. Atthe end of each ofthese branches, there are sates of nate nodes
‘aos of nature emerge as sub-tanches from these nodes. The prota
(Stein fom th et ih od ft enon sop peta
off cch ours nd en the eu fs a pe mem
Seva for ich fhe rcs. 0) red backward the erage oF
Sesion pt
Repo the pil he ist econ nodes False he
Scum otacon ok lsd om the bepming ote edhe te he
sata rion ene wtale
‘esha ose ther dvnn probe sng esos ws apprench On
tech mawofestus neh wie hese posh wd yoa4 Business Saves (FYB MS: SEMED
EMY at(1)= $000 + 0.6 + 3000» 0.4 =F 42000-
EMV at 2)= 11000 06-4000 « 04 = 5000/-
+ Aon; nin ft pian ea ae sr
‘Product, =
mactration #
Consider the illowing decision making problem.
Payot Table
Course of ‘Sate of Nature
Action % &
a 25000 | 35000 | 40000
ws ‘son00_|_ 20000 | 10000
Frabbiiny | 03 03 o
Soteion +
The fllonng isthe Decision Tree
Decision Theory as
CCaleatation of EMV
For dy EMV ~25,000% 03+ 38,000%0,5 + 40,000 = 0.2
500+ 17500 + 000 = 3,000,
FocdpEMV ~50,000 033+ 20,000 0.5 + 10,000» 0.2
15,000 + 10,000 + 2000=27,000
AS EMV is mare (3,000) for ation 4 the eson
‘more EMV of€ 33,00,
Iasteation 9:
"Nature Cream Lid wants intradve anew brand ofie-reams inthe marks.
The follwing alert re availble tothe company.
Open Five New Outlet: In his case ifthe demand is high profit of &
sh 000 pm. iseapeded Ifthe demands medium, profit ante oF 50,000
tnd low demand ican be® 30,00 ool:
(i) Open Ten New Outlets: nhieu the proftecente® 10,000, 1,10000
‘2818 60,00 fer demands beings hgh, peu and low
Pas experience shows the probit ofthese demands a 0.25, 045 and 0:3
eesti Sugger the opal decision forthe company using Deision Tree
Method
to hose ation 4, with
Sotaton
‘The given information ca berepresnad as pay-off. Let SS, andS denote
the sacs of ature correrpanding to demand beng high, medium and Low
‘Ropes Smiley le fy beth conse of action opening Sand 10 new
lt espe
Pay aff Table
Coane State of Nature
Action x 3
a 0,000 730,000
be 130000 60,000
Prsebiniy [025 03
Catetation of EMV
Fora, EMV =025 $0,000 + 045 » 50,000 +0. «30,000
= 20,000 2,500 +9000 = 51,500
For dy EMV =025 150,000 + OS « 1,10000-+0.3 » 60,000
37,300 49,500 + 18,000 105,000,46
AG EMV ie more for ation 4,
Busines Stasis EYB.M-S.: SER
‘pening of 1D outlets wih EMV’ of 1,050
Mstration 10:
£5 00), the desson 0 choot ation
whose returns ar sed on ste of economy, Posse ste fn
Nett anon ee sng into pana
‘conom)and associated pcb distiuton ae given low
as flows
BacoNewre| Far [Gad [Gre
Probability | 02. oe 3
‘Theremrasin Rapesforeach investment cppormitysndeach a ofnatremm
“Bieraatver Baie of Feo
Fair [Good | Great
& woo | 3000] — 6000
a soo | 4500 | 6800
an ° 3000 | 5000
a -s000 | e000 | 4500
Constr eison oeand ogg the bee aharnative or van
Decision Tory a7
Satution:
The fillowing isthe Desision Tree
[EMY Caleulation =
oc = (0.2 « 1000) + (0.5 3000) + (0.3 » 6000)~3500
Fe d= (02 » $00) + (05 > 4S00)+ (0.3 = 6800)= 4390
Fo dy = (02 0) + (05 500) (0.3 8000) = 4900
ax y= (0.2 * A000) + (0.5 6000) + (0.3 8.500) ~-4750,
‘As maximum EMV = 4900s for yp
Optimal decision ito choos alter Ay
lestration 11
‘A has ficory specialising neta i developing sustantlbacklog an i's
management ccosierng three ovaries of aston. Arrange for subcontracting
(ss begin overtime production (S,) and construct anew facility (5). The crest
‘hole lrgely depends upon fire demand which ay be law, mer or high
Byconsensie management ranks tie espe probuites 60.1, 0.Sand 04.8
cunt anal reveals effets pen the profits as shown in the table belowae Business Series EXBMS.°SEM-IE—pecton Peory ae
‘Aivernaties| ‘Sate of Reonomy Coarse of |__Stateof Nature (Demand Level
7 Tow | Medi Ta ‘Action [igh | Medio [Low
7 10 0 30 Taree Pant | 100 @ 20
s 20 60 100 SmallPant | 25 3s as
5 =i | 200 Pratainty | 0s a 02
Constr the dicison ice sing above information and gas at course
Solutio
The flowing isthe Decision Tree
EMV iscaelated a lows
(For Aterative Sp EMV=0.110-+0.5 50+ 04» 50=46
(Gi) or Ateraative, EMV= 0.1» (-20)+ 05 (60) +04 (100) = 68
(Gi) For Atternative EMV =0.1 x 150) +0.5« (20) +04 « 200)
5
‘The best cour o tion iso choose allemativeS, wih maximum EMV 75,
tsration 12:
nese AE pues
eer eee
aera
eee, seer
lakhs of) over the next 10 years, = —
(Comsrut decison re and determine the best deciton wing EMV mated
Solution:
[Net Payotf= Sales — Cost
EMY Calculation
For Large lant = (05 15) + (0 38) + (02*-5) = 847 ake
For Small plant = (0.5 « 13) + (03 23) (02 38)~820 aks
By sutractng the respective costs of 25 ac an 12 las,
Net value for Large plat = 47 lakhs 25 ths = 22 lakh
‘Net vale for Small plant~20 lakhs ~ 12 aks ~ aks
2. Optnal detsio sto build Large plant
Iiastration 13:
A manufacture o ois interested to know whether he should launch a dele
‘model, luxury mos or popue modelo toy. With the dele model the
efits will be€ 70,00, € 35,000 and ® 50,000 wit probabilities 03, 04 and 0.3
‘spective Fr alunury model, praia would be 60,00, 40,00 and 55.000
‘th protabities of 0.40.3 and 0 especie Fora populate patie
‘ill te € 40,000, € 30,000 and € 28,00 with protabites of 05, 08 and 0.2
‘espetively Draw a Devon Treeto determine which model shuld be launched
‘omanis isp2 70,000
F350
50.000; EMV =sa000
£60.00
£40,000
55,000; EMV = $3,800
4000
30.000
£25,000; EMV =34,000
‘Launch Lanury mel with maxiaum EMV of 52,500
EXERCISE
2 Amamge haste sh ferries fa
ini Turow pie saes nace doa:
xan a ed 9 openly Peas Aer
uae oN snopes ea
Path wet 00, 38,00 de bee
iio wih the lo Den ese ss
Sinden pee
Primes Sess (FYB. MS: SEM
ection Theory am
Fs
2.
EMV esieulition are 38,5007, 39500, 24500, Hence best decision isto
choose ation 4, with EMV 39,500,
‘person has to make a choice of purchasing 100 shares of Conpany foe
{Company 5 the markt is high it will ox him 28,00, ti fale
‘illo him & 25,600 and fer alow market it wl est him 220,000 oe
company 4 While fo company, the cresponding amount are® 27000,
£25,000 and 26,00 repetvel or abigh ian low ret condo,
‘Therespeciveprbablites or hese marke conics we 0.3, 0 and 03.
respecvely: Draw approprite Decision Tres and advice the person about
purchase of hares,
‘The expcted ests are 24,40 and 25,700 Hence iis visio purchase
‘ares of company A with less ost of 24,400,
‘Aconnpanyhat to decide about sales fnew predac among thes regions,
‘and I. The estimated sales fates in region Tate 2,000 unis, 18,000
‘nits nd 1,000 units tbe demand is igh fran lam respectively, Foe
‘erion theses garesare 21,000 units, 19,500 its and 14,000 aso
these demand conensrespativay. Wheres, the exiles of aes ae
2.4000 units, 14,000 uns aed 2,000 unt respectively fe tere domatd
statis i epon I. The probabilities ssc with he demand being
‘lg, tar and low are 0.25, 048 and 0.3 respectively Suggest the tial
refion using Decision Tree Methodsa Busines Setiticr FYRMS: SEM-D
2000
‘The expected eles ar 17,600, 15,150, 15,90, So best region i elt
with maximum EMV of 1,600 is,
25. A company has the opportunity of marksting # new package of compu
ames. Ithas wo pouil courses of action tert markt on mite scale
{o give upthe rjetcompleely A test market would ort &1,60,000 aa
‘curent evideice sugaesitha contre facing ikely
egies Ithe reaction test marketing ix poi the company
either market the computer games ational o wil give yp the po
Ecenplcy
Research siggeats thats atonal iauach might retin the lowing
‘Sales | Conteibation © mations) [Probab]
igh 12 02s
‘Average 03 050
Low 024 025
‘he test marketing were yield ‘negate rents, the company mould
‘pth project. Giving up the projet at any pit would resin conti
‘F € 60,00 from the sles of copyright eo note manure
enrbationsbaveteen discounted to present val.
You ae resuired 0
Decision Theory a8
(©) Recommend cose fain forthe company on theasisofexpoed
ans
/— civ pom
eo
2
oD 2
Marketing, is
05 45
"$024
«025 +033 «05-024 4025
= 039 milions “€3,90,00
td giving up the projet wl ch € 60,000. So if market ection
‘estive wil be beter o produce
(G)_EMY fr test marketing is 0.5 «350,000 + 0.5 60,0008 2,25,000
Net gin fr est marketing = 2,25 000-1 60,0 ~ 5,00,
‘mhich more than giving up gain oF€ 60,060. So best eiton ito go>
atest arkting.
26, Draw Decision Tee forthe flowing problem and suggests best course of
Action (Use EMV Criterion,
(BMY ot produtien
Choise of ‘Raters of Demand
Product Fur | Good | Bow
4 ao] 290] s00
a wo_| 40 | ooo
Prstebiliy Demand) | 02 os a7
Business Sanies (YAS: SEMEL
[BMY for A 355, fr B = 480: Choose Produ 5 with wena EMV of
a,