PG WEEKNAPPLIGATION DEVELOPMENT

MAY 1, 1989

ME ia pe a a —ni TT a urea id -a

APPLJED INTELLIGENCE.

DSS Applications Should 5hed New Light on a Problem
pučejs The diversity of functions within DSS general DSS or business-modeling tool. programming skills are often Ee in
isi tw roducts makes it difficult to group all As described last week, executive in- order to make effective use of the |
Thas ts part (wo ; roducis into one category. By un- formation systems are specifically de- language. | SEM |
a dei s ding the needs of the. users and signed to help executives gain insights The focus of business-modeling prod:
abeje ai oe pa GA poč needs into software and track critical success factors. The ucts is on a robust modeling facility
tive tnformalon na GA E there oken five distinct focus of an EIS is to aid a decision mak- | with analysis capabilities and on tools |
syslems ana de- — lune osat. support products: er in assimilating information guickly to transfer data into the modeling envi: |
else teci A rea daleet products, EIS, business-mod- and identifying PN ina EH tuni- Min fr Vi kteri |
nologies that en-  eling products, general DSS products ties, not as an aid in problem analysis T  EPA/PIE (romi Brešan STE |
able erecutives and statistical-analysis products. or resolution. In many corporatlons, ATE US, ora ača |
im di Of the five, spreadsheets are the most. well-developed EIS have replaced the Corp., and System W, from Com |
klana Pipe beg ded A SEMENA enables us- . traditional periodic executive summary The fourth ES z sie zalijejo
mation. ers to perform the first level of analysis reports. Representative products in this DSS product. 1 se ka luc : ar | |
Decision sup- and to develop simple models. category include Command Center, from to but more powerfu ea pkojelisii š
(k Ri Hi | port systems Typically, spreadsheet data is not Pilot Executive Software, and Com- modeling tools. Among the distinguis
MARTI (DSS) can be accessed through a data dictionary; the  mander EIS, from Comshare Inc. ing characteristics of general DSS prod
tedn Business-modeling products are used ucts are the facilities provided to man:
age multidimensional data and the ease

considered user has the freedom to perform any

a RATIO STR NTI TE 2... | in which data can be added to the sys
| concepts that are either hard to grasp ki | ii
| or are constantly being redefined. Sim: |. ...%;:/ D nici n Su ort S stems | tem. These products are intended to
| ply stated, a DSS is any system that can | Atena CCISION LI pp y eš SEE na ULE
be used to increase the efficiency and ; Serie | . P s nui.i..n: . roader set 0! user needs.
effectiveness of the EIN pro- Let Managers Examine Data in Multiple D 1mens10NS These types of DSS products also pro
| SRAT ee UE Ar AIR O VE srdeeko ME RA AM (ii vide facilities for creating highly format:
1, Yet, the concept of a DSS remains: ' Using decision supporttools, |: .. ...orrevlewsalesdatain | ted reports and a variety of Brap Le dis-
|| Ag many people. To Sara plak Skeeen hi zesales " .. . multiple other dimensions... | |  plays. A Very ZOE obi ura! lan:
means end-user computi rsona »- in terms of product lines... /;i-.." Hp SN Se nji je guage is provided that reguires
— -computing, or s ee vrtele kj Sys- lisic, kia O RE HA SN dode iSales PETE lje RE Enej skill for effective use. The lan-
tems (EIS). However, these terms are Š : Techligu6a my Review... |  guage can be used to write user-oriented
not synonyms for DSS. (NET SEE SESIE ii i emi (V Salesby " | procedures such as screen data entry,
The general characteristics of a DSS O oe a NE teview MM calos ba ka: ustomet | processing and reporting. Representative

j
>

a RR EN OE NV
iki

s" sa ei NE KE pi EI i
o eo rea, Voda
AE Aa MetEh

kajo, |
Pod de

i

-.

"-- k a RE
"a, — — pe

x

ii
it
"

ae

MI k

ksi a
P
peč 24
O aja?
laki

E

z

i in a DSS is more aggregate and less from Information Resources Inc., and |

stored in the form of multidimensional (Er nee

ani accessible to a non1S professional. mg Gel | used for the statistical analysis and

nities. Where a general DSS product may

er applications. As corporations succeed- including linear, multiple, polynomial
additional reporting reguirements of derstand the appropriateness of the |
ty, not just reporting, to support their i K SEE Rai NSi, — analysts. SAS, from SAS Institute Inc.,

met by DSS products. ; son Micro products are now beginning to

record-oriented than information sys- sle BE |tONeEO 44/11 |  Stratagem, from Computer Associates
Review Marketing ' | | ŠE
| the 4 mi ng The last category, statistical-analysis
arrays. The interaction between the DSS
Sukič IE ra rr mra
The DSS is gene identif 20 SSglpč |, tistical routi
| generally used to identify Orpotale dica) >; |  Ustical routines are more extensive than
Decision suppo stems evolve« za MA € B Klej ud TRKA Sre! ji li moč | :
| stat uda A, š Tem Ti Ev MM 5 Ra, zom have one or two types of regression, a |
ed in automating well-structured opera-
ne and stepwise.
those applications. Users (analysts and RJZA NAV
| technigue and how to interpret the re-
hat provided flexibilitv and fimetianali. — |." 4
that provided flexibility and functionali — , Product o rn SE (Z | sci ih
. . . . pire 1%; ge AM: : di nih zač: pasla ude sj oe. ši PARE seli ni .: Js z: ii . ku zni he ua < m sito: č Fi t | | | |
problem-identification and decision-mak- L' ' .: s kO NI SEEN MIMA nena ii pd: zen 89 Hal and SPSS-X, from SPSS Inc., represent
One important characteristic of DSS | provide capabilities previously found se

distinguish it from other systems. Data zirali products in this category are Express,
tems (IS) applications. DSS data is z
Assumptions ; | |
MO zNAa, nike, products, consists of products that are
and the user is very flexible, immediate iz oniaj
i TRTE Pie |e Ee VALUTNA < ——): | re h . .
A a rl Ee SI modeling of data. Their libraries of sta-
and understan blems ; rtu- Melej pe ji | a raja pe ote ZE TRR O
d problems and opportu RE pe i see Data those found in the general DSS products.
gradually from more structured comput- nie
p statistical-analysis product has several,
tional systems, they began to tackle the
Generally, users are expected to un-
executives alike) began to demand tools |.:.|1]|:/ sika
ta STER sults. Usually these are used by trained |
š (88, NP
ing needs. These evolving needs were products in this category.
products is their ability to support a only on the mainframe. Mainf rame DSS

pa sa model and retrieve data One important characteristic of DSS products is their ability prolwes are o kaated tothe mlera TERIM
managetneni systems (DBMS.) as veli. O SUPPOFL a potcerful data model and retrieve data from | be espanded anid enhancce, paj L
support structure must din te pro ooo bodalo" DBMSS as well as from external sources, Moro $a cah Fun slandalone on the. O,
( cesses of ad hoc dueries and analysis, s; Em ne —— There is no doubt that ea ea IE
č co DBS will be further modified as the mi. o | |

and must be flexible enough for the user A

pi
m
V

data as the problem is operation, logically valid or invalid, on by financial analv |
ze m , data in the spreadsheet. The lack of ta a, o vas lo represent rela- — revolution coin
EN AN, kri metan ph many ponskips between dala elements in the. Next mes, well ok at Um Ze
ihe an-. anagers to become wary of informa- - forward as a profit-an Šk iona nav Managera ihould look for in a DSS
tely for- tion beige im Kaka para Ja statement or a balance sheet. More com- IBanizatlonsa ———
lon, Spreadsheet facilities include func-  plex busi nes modeki zele Com: SM

Hb rok dij ik
ši -— Mmm
| NES