




























































































Study with the several resources on Docsity
Earn points by helping other students or get them with a premium plan
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
The slow progress in improving the front end of converting systems from manual to machine processing, specifically in the areas of systems analysis and design. The author argues that a new framework is necessary to effectively perform systems analysis and design in newer areas. The document also explores conventional systems analysis methods and the use of models in the analysis process.
Typology: Slides
1 / 106
This page cannot be seen from the preview
Don't miss anything!





























































































A Methodology and^ Case^ Study by John F. Rockart September 1969 Working Paper No.^ A15-
50 MEMORIAL DRIVE
DEWEY LIBRARY
A Methodology and Case Study by John F. Rockart September 1969 Working^ Paper^ No.^ 415-
r]o, ^-, C.3.
T. (^) LIBRARIES
More recently, Canning has pointed to the "painfully^ slow^ response"^ to management needs provided by the current techniques of systems analysis and design (3). Nor are these^ two^ alone^ in^ their^ comments.^ Words^ like "cumbersome tools" and "the tendency to simply^ copy^ the^ old^ system^ over into the new" sprinkle the great bulk of the literature^ on^ the^ subject. Today, most of the more common traditional programmed functions of^ bus- iness management, such as payroll and order^ entry,^ have^ been^ already^ trans- lated into computer processing. To handle new and^ less^ well^ understood functions, which^ are^ now^ of^ greater^ interest,^ the^ old^ techniques^ of^ systems analysis and design are sadly lacking. In^ order^ to^ effectively^ perform systems analysis and design in these newer areas, a new framework is necessary. In the area^ of^ systems^ design,^ model-based^ systems^ design^ as espoused by Carroll (5) and others appears to be one^ key^ to^ innovation^ and greater certainty of results in the fabrication of a new system. The step which precedes the design^ step^ in^ the^ data^ processing^ conver- sion process is, however, also in need of a new framework.^ It is^ this initial systems analysis^ step^ (sometimes^ referred^ to^ as^ the^ "systems study") with which this paper is concerned. After a discussion^ of^ exist- ing conventional systems analysis theory and tools, a three-pronged, model- based, systems analysis^ theory will^ be^ presented. First, let us pause to establish some definitions.^ In^ general,^ the process of converting from a manual system to an automated on has^ been divided into three steps, systems analysis, system design, and programming.
by Simon^ In^ the(4)^ sense^ of^ programmed^ versus^ non-^ programmed^ functions^ as^ described
.
-3-
"Analysis ... is finding out what is to be done; design is finding^ out^ how it should^ be^ done;^ programming^ is^ making^ the^ specified^ system^ a^ reality."^ (6) The distinction, as performed^ in^ practice,^ is^ not^ as^ clear^ cut.^ "As^ we know, the people called systems analysts actually do some of the^ work called systems design and, in some cases, so do the programmers." (8) There is extensive interaction^ between^ phases^ of^ the^ process.^ However,^ the^ analysis function, no matter when performed, is clear. It "is restricted to fact- finding and to examining systems to learn how they work ..."^ (8)
Conventional Systems^ Analysis Conventional systems analysis is described in varying^ ways^ by^ different authors. (^) (9, 10, 11, 12, 13) The process, however, is in general agreed upon by most as a series of steps dominated by interviews of operating personnel and data collection in the area to be studied. Typical of these formulations is^ a^ five-step^ program^ for^ the^ system^ analyst^ presented^ by
"First, obtain facts by interviewing people and observing activities timing-- that about lead^ theto^ events--theirthe origination^ type, of documents,volume,^ and main- tenance done at ofeach files, work issuancestation, ofand reports,flow of processingdocuments betweensteps stations
.
"Second, collect sample^ copies^ of^ filled-in^ documents,^ ..." "Third, every document study processingthat each operationsperson receives to learn or issues..."the how and why of "Fourth, organize the facts obtained into flow charts, flow origin,^ lists^ or throughother^ suitableeach stage^ form of^ tocommunication^ trace^ the^ path and^ ofprocessing,^ data^ from into files, and out of files to reports. "Fifth, interview each user of documents and reports to learn what needs." information (14) he^ uses^ in^ his^ work^ and^ what^ he^ thinks^ he
Exhibit 1 Input-Output Chart
ORDER processing^ Input^ and FOR^ outputSTOCK^ ITEMSof^ customer ^SEMSm
Information^ Reproduced Systems,"^ from^ Evans,Harvard^ M. Business^ K.^ and^ Hague,Review^ , L.Jan-Feb^ R.^ ,^ "Master^ 1962, Planp. 98.^ for
The use^ of^ I/O^ charts^ leads^ to^ an^ identification^ of^ the^ significant^ data items in the particular system being analyzed.^ It^ allows^ the^ elimination^ of redundant data item inputs. Finally, it focuses attention on often-used per- manent types of^ information^ which^ should^ be^ stored^ on^ a^ master^ file.^ (17) A. Statistical Sampling Techniques^.^ It^ has^ long^ been^ recognized^ that^ a relatively few well-chosen observations will permit inferences to be drawn re- garding the total population^ from^ which^ the^ sample^ was^ drawn.^ In^ order^ to reduce system study costs, these techniques^ have^ necessarily^ been^ utilized.
tool, but it is often used.
Data Presentation Techniques