THE TASKS TO WHICH THIS STUDY IS ADDRESSED ARE THE DEFINITION OF MANAGEMENT

• Document Type: CREST
• Collection: NGA Records (Formerly NIMA)
• Document Number (FOIA) /ESDN (CREST): CIA-RDP73T00325R000100010003-9
• Release Decision: RIPPUB
• Original Classification: S
• Document Page Count: 13
• Document Creation Date: January 4, 2017
• Sequence Number: 3
• Content Type: REPORT

Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
tiiv}.ICii~u F~ti LR ~ S-L-C��}'~-i -T
Dtain Re~orf.:
ir'JR}~iiyv Prl,'-'~ E
The tasks t0 t5'hich this study is addressed are the definition of management
i.nf_oxmation noels, reassessment of the currant Management Information System
(I~iIS), and consideration of alternatives. The intent is to vieti~~ the MIS and
management information needs, and their relation.shi_ps to Center activities, in
general, in broad perspective so as not to obscure the issues titi~ith unnecessary
detail. No attempt will be made to fully summarize the appendices, since that
information speaks for itself. However, the appendices will be liberally
referred to and significant points noted.
It is extremely difficult to place a specific value upon management infor-
mation, particularly in anon-profit envirownent. ~'lbout all t}.at can be said
is that Center management (and other appropriate authority) should have avail-
able, by one means or another, that information which S.t must have to accomplish
its objectives and fulfill its missions and that the associ.atcc}. oasts should
' not be excessive. Certainly this information requirement ca.n be met at a cost
commensurate with the importance of the NPIC product and with fulfillment of
the national role in the intelligence con~nuni.ty.
The problem then is briefly one of, firstly, defining the objectives and
missions and, secondly, defining the information which is essential to accomplish.
these in order to assess the assets and liabilities of the MIS and consider
alternatives. '!he importance of objectives, goals, etc., has been discussed
clse4ehere a number of times, so the focus of this study will be upon the second
item. This deceptively complex subject, the information necessary to manage,
nof.}iing more and nothing less, is the legitimate concern; that is, a management
information. system, not a strategic planning tool., not an operations researc~i
~ technique, an c}-e_~initely not just an operational control device for monitoring
specific tasks, although this latter activity is certainly related to a management
information system and may indeed. be a subset of such a system. ,
There is much confusion about management information systems in general and
the MIS in }?articular; this is the root cause of many of the problems besetting
the current ~1IS. Questions such as "bV}lat do I want it to do and why?" "tVhat
should it do?" "tiVhat can it do?" "iVhat can't it do?" need ans~~=ers. Stated
bluntly, it is Quite clear that many persons do not either understand or agree,
or perhaps both, upon the purposes and capabilities of the MIS, or any manage-
ment information system for that matter. This is evident from the responses
obtained from the Groups and Staff, and from discussions ~,~ith IAS, to certain
Questions posed to them on the 14IIS and management information needs (see
Appendices I3, ~ues~tions on Definition of h?aiia,~2_ement Information Needs and
Reassessment of the T�4IS, andand E, tir-ii c ~ surrm~arizes responses to the Questions) .
Perhaps, too, it is naive and unrealistic to expect full understanding and
agreement, and it should not be inferxed that any one answer to a. Question
posed is necessarily correct. The answer to a Question such as "lvhat are your
S-E-C-R-E-'?'
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
� �
. ~~~RKiA1G PAPER
Main Report
d
S-E-C-R-E-'I'
~t~JRKINC PAPER
management information needs?" will depend upon, among other things, suc}~
variables as what the person is supposed to do, what he actt~a.lly does, how he
does it, the particular problems he has encountered v1 the past, the requests
he has had and anticipates for information about his areas of responsibility,
and even personal biases.
When trying to categorize and analyze the substance and validity of responses
stating information needs, plus opinions as to what is wrong with or lacking in
an ongoing system, which also reflect the variables mentioned above, it is next
to v~ossible to classify, much less reach conclusions about, the responses
unless some sort of framework is provided upon tivhich to ''hang" the information;
that is, to relate its parts and make it a cohesive picture, so as to furnish
a perspective which is valid and can be of some use in making the appropriate
decisions.
Such a framework is provided in Appendix D, Management Tnformation and
Planni~n~ and Control. Although some might consider suci an approac "too
t~eore~---tick.'' anima waste of time," it is hoped that the point has been n~a.de,
and it will be reinforced below, that this is absolutely essential. All too
often, there is a tendency to go charging off in a erections with solutions
to the problems before the problems are really defined or understood. The
intent herein is to derive the right solution for the real problem, not what
someone thinks it is but what it actually is. (For example, would it not be
rather foolish to make the MIS real time when there is alinost unanimous agree-
ment that the btIS information content is inadequate?) And it does not matter
so much whether or not there is agreement upon the cor..ceptual framework (there
probably won't be since there are almost as many views of the management process
as there are managers); what is importaaxt is that the classification schema is
understood so that there is con�~munication using a language meaning the same
things to all involved. For example, it is hoped that, when operational control
decisions are referred to, all are thinking about roughly the same type of
decisions with respect to tune span, organizational level, importance, amount
of judgment involved, and so forth, no matter what others might have preferred
or even what might have been preferable to term that class of decisions. The
main definitions derived in Appendix D are repeated here for convenience.
Strategic Planning: the process of deciding on objectives of
`r t}ie organization, on changes in the objectives, on the
gross resources used to attain these objectives, and on
policies that are to govern the acquisition and use of
these resources. `
i
,{
Management Control: the process by which managers assure
t resources are obtained and used effectively and
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
� �
. 1~'Oi2hING PAl'L'-R S-E-C-R-Ii-T }ti'ORKING PAPZ12
' Atain Report
efficiently in the accomplishment of the organization's
objectives. It combines both plaiuiinn and control within
the context of objectives and policies deternuned in the
strategic planning process; effectiveness and efficiency
are the criteria relevant for judging actions.
Operational Control: the process of assuring L-hat ~s~~ec~if~ic
tas~are carried out efficiently and effectively- �, emphasis
is upon execution of tasks.
It was not possible to obtain the first or original cost of the current
MIS, but a very conservative estimate of current aauiual costs to NPIC, more
or less directly attributable to r~1IS operations, i or, looking at
it another way., less hardware costs of about yearly, this is the
equivalent of about 1 NPIC average gra e mars-years. It should be
25X1
25X1
25X1
noted that ,hardware cost is essentially a sunk cost since NPIC os,ms its computers
and contracts for regular maintenance, and those costs equivalent to the extent
t}rat MIS output obviates other systems probably should be subtracted from the
annual cost figure if it is desired to arrive at an absolute cost figure.
Appendix C, Bxief Description of the Cuxrent NiIS, and Appendix F, Samples
of A~fIS Standard Output, roughly sketch the operation of the System and indicate
its capabilities, Briefly, the MIS can pro~ride regular time hours, overtime hours,
and dollar equivalents sorted in various ways according to organizational element
(Group, Division, etc.), project block (major work effort, suc}r as Y~}-I-4A
processing, basic reporting, etc.), specific project nwnber, and activity or
skill code(s). It is capable of providing reports or lista.ngs showing active
projects on record, active projects by responsible component, all projects on
record, netiv projects established, projects completed during a period, projects
~ cancelled, and products and time recorded against cancelled and completed pro-
f jects. These reports generally include project number, project title, organiza-
tional element responsible for the project, category or specific type of project,
cowitry or area of interest, requesting organization, project priority/weight,
1 begin, deadline, and completion/cancellation dates, and security classification
' code of the requirement form. Also available is a report, Component Time
1 Allocation, giving weekly information on each employee by component; it shows
name, badge number, grade, component, week ending date, projects worked on,
activities against projects, and regular time, overtime, and dollar equivalents.
I In addition, there i.s a A-iIS capability for specially formatted reports; that is,
presentation of data in other than standard formats in answer to ad hoc requests
but not necessarily in any format desired--there are constraints as to how
the data may be aggregated and presented.
-~
i
3
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
i;~RKING I'APl::R � S-E-C-R-E-T i~JRf;ING PAPLR
Atain Report
In short, this is the current D�iTS capability, cvhi.ch does not meant t}~e
System carniot be unproved, i� desirable, in a number of ways, including perhaps
reprogramming. 1'he most serious constraints by far on improvement of the
MIS are dictated by its record structure, which simply means that it is
possible to input and store data only in certain rigid formats and amounts and
to associate the data in only certain ways. In other words, there a_re some real
� constraints upon what may be done,~within xeason, with the current MIS. By
"within reason" is meant that, if we are speaking of making the b~IIS into a real-
ti~ne system, with near real-time input and outp~at, then, for all intents and
purposes, we are, or ought to be, speaking ,about a new system, one in which
� it ~yould be worse than foolish to incorporate these serious constraints. This
implies a very fundamental change, comprising redesign and at least major re-
programming. If, on the other hand, we are really considering only malting the
current AtIS configuration real-time; that is, imagine t}ia.t essentially the
current output were available in zeal time, then we are no longer speaking of
a management system with real applicability to management planning and control
(excepti.on reporting), but only of what would be basically an operational control
mechanism.
The current MIS is mainly oriented toward project accounting, or the re-
cording of data against discrete tasks. This has meant that, since all the
time of every employee is recorded in the MIS, when utilizing the MIS all Center
activity must be conceived of as being applies} to discrete tasks, an obvious
and serious error. The System tries to "hedge" against this misconception,
which unfortunately has had certain ramifications in holy some persons fundamentally
view the activities of the Center, by setting up "open" project numbers and
other devices which in fact obscure rather than aid assessment of accomplish-
ments and often greatly frustrate management attempts to use the System. Perhaps
it is of relatively minor importanee, but this has also increased costs through
undesirable redundancy between organizational element, activity code, and
project number.
Not to belabor the point, but this essentially task-oriented approach is
conceptually invalid and is the greatest single detriment to the System's
imputed purpose, to provide management information. A1.though there is some
legitimate management information "fallout" fmm the System, comprising certain
aggregations of project oriented data, the System is by design basically an
operational control mechanism which has been forced to masquerade as a manage-
ment infoxmation system. And as a specific task-oriented system, by consensus
it fails a most crucial test--it is not timely enough.
~~~9
1
In managing specific tasks, it is important to know when, and also how
far, something is deviating from plan or is out-of-control, so corrective
action can be taken eithex by supervision or line management. It is, of course,
4
S-E-C-R-F.-T
' Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
l;'OR1~I;vG PAPEic S-E-C-R-E-T }VORKING PAPER
Main Report
,.
a matter of judgment and a. function of the particular activity being monitored
just how timely inforniation musi; bc, but apparently the System is not timely
enough because it rarely, if ever, is actually used for control purposes by line
management and supervision; that is, either to ascertain if things are going
according to plan or if things are out-of-control. The Pi41B is moving into the
control area, monitoring projects using the 1`tIS, but its position also is that
t}ie MIS is not timely enough for this purpose. Acid with respect to monitoring
specific projects, upon which the P~tL3 scheduling function is actually based,
one could tivonder if this line function, project scheduling acid control, has not
been relegated to what could be considered a rather high-level staff activity
(even though the P~~IB is composed primarily of line representatives who meet
daily) due in great part to the failure to provide line management with the
data it needs in a timely fashion. Of course, the P;,-1B has the additional
responsibility of scrtrtinizing and validating rec{uests for products or services.
As mentioned, it is the consensus of NPIC components that the ItitIS is not
timely enough for those uses the components envisage for it. It is safe to
. say that IAS would prefer more timely output. As pointed out in Appendix C,
output relating to time sheets is normally received about one week after the
' data is submitted to AID. This can mean that, for a specific project, since
the time sheets cover a one-week period, it could be up to ttivo weeks before
information were available; for example, if the information pertained to time
recorded against a project completed on a A4onday. For other types of data,
even though the data is often input daily to AID, the lag in time can be from
about ttvo days (as a best case for items such as products produced, changes
in project status, picking up new projects, etc.) to about one full week.
This is because ulxlating is accomplished only once a week.
{ No system can be all things to all people, in spite of what some computer
specialists will say, but this is the position in which the MIS has been placed.
It is.perhaps worthwhile to note that, while system's designers and operators
may have many virtues, managerial insight and competence are n.ot necessarily
among them, but management is often afraid to challenge these specialists
publically lest it display its ignorance. Such discussions usually get bogged
down in detailed technical considerations, freely laced with confusing computer
jargon, and the battle is fought on the computer specialist's ot,~i field, so to
I speak. This is unfortunate, a~~d it should be just the re~rerse--t}ie system
exists to serve management and the computer center is a service organization
which should be able to communicate with management in the lanb age of manage-
~ ment.
Because expectations and needs have not and can not be matched by per-
. formance, the btIS has failed to do any job as well as it should be done, either
in the areas of strategic planning, management control, or operational control.
'"I It might be referable to modif as. needed and im rove the btIS
b p y, p (incidentially
5
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
� titi'OP.ICING PAPER S-E-C-R-E-T
Main Report
1`~RKING PAPER
also to rename it) and essentially assign to it the job for which it is bast
' suited; that is, monitoring specific tasks in the operational control area.
To expect it to function truly as a true management system, at least in its
present configuration, is just unrealistic. Perhaps with modification, ulti-
� mately it could have real utility for those discrete activities which are
' amenable to establishment as specific projects. This does not mean, however,
' that the management information "fallout" mentioned earlier should be disregard-
ed unless or until other provisions are made.
Interestingly, the MIS has found its greatest applicability in an area
i.nconsistent~with its design; that is, in the area of what is primarily manage-
ment planning (a subset of what is called management control), This is exem-
plified by its use in the PPB process, including PPB documentation; for example,
in the planning and programing of resources to accomplish approved program
elements and sub-elements. And within this use, the non-financial accounting
function is greatest; t}iat is, keeping track of how manpoeer resources were
� applied to major work categories. Rough measures, based upon this historical
' data, are used mainly for projecting future manpower needs; those who use these
measures are suite aware of the slack within the System, The total financial
aspect, the touchstone of the PPB System, is essentially outside the IvtIS
capability, as is any object class/program element crossivalking, In addition,
certain other uses are made of MIS data, such as forecasting for the Pit~tB, with
the express knowledge that the figures are rough, overall planning inputs wherein
' the values of the real variables comprising the estimating statistics are sat
~ known.
Aside from the obvious deficiencies in capabilities and operation of the
~ System., too much information needed by line management is not incorporated;
i.e., financial, human resource, and supplies and materiel resource information,
r especially if film with all its attributes (footage, targets, schedules, etc.)
is thought of as a materiel resource--perhaps an uncontrolled input, but none-
theless a material resource. In the future, electronic transmissions of a
l real-time collection system would be analogous to this primary materiel input.
~ Appendix D makes a point of the primacy of the management decision process,
whether it be for planning or controlling purposes, or any blend of these
activities; this is the area toavard which a management information system must
~ be directed--it must provide comprehensive inforniation on the ongoing operation.
~ To be comprehensive, it must provide human resource, supplies and materiel
~ resource, and internal and external financial information geared to the type
of decisions which must be made and to external requirements for information,
? and all this must be aceomplish~d in the proper time frame. Also, as was noted
in Appendix D, the central f~mction of a management control system is motiva-
tion, motivating management to make decisions and act in t+fays consistent with
the overall organizational objectives. The relationship of the system to
other systems must be recognized, but the system should filter and condense
' ~ 6
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
�
jti~Rt;ING PAPER - S-E��C-R-E-T
A~tain Report
tivOR.KII`~ PAPLR
information, removing that which is irrelevant for management purposes.
While the MIS is used primarily as a historical record, often the basis
for rough manpower projections, and relatively little for operational decision-
making by line management, in those cases where MIS information conceivably
could be used for decision-making, there is a great deal of hesitancy to do so
because of deep suspicior~ concerning the accuracy of the information. (There
are exceptions concerning accuracy in some components and the IAS.) The accuracy
concern basically is a function of three things: (1) certain aspects of the
System itself, not amenable to change; i.e., it is not sufficiently reflective
of Center activities; (2) certain System mer.}ianisms and procedures :vich frustrate
accuracy of input and output but which are amenable to change, and (3) a lack of
interest in the Systan resulting in little enforcement of the procedures,
inadequate validation of input data, and inattention to potential uses and
worthwhile improvements. Management at all levels must share the blame for
the latter tsvo problems; it generally does not "back the System" and turns a
deaf ear even when real problems are surfaced. Of course, this is not true
in every case, and those components in which it is not true are precisely the
ones which derive the most benefit from the MIS.
Further, this lack of backing is not necessarily the fault of top manage-
ment, as is so often charged, because it is the responsibility of the lowest
level of management dealing with the MIS input data to insure that it is accurate
. and as timely as the System permits. The Executive Director, for example,
neither can, should, nor will check all t'~e time sheets for accuracy on D9onday
mornings; all he can say is that the input data must be accurate and the Group
and Staff Chiefs must reflect this command and pass it on to the appropriate
~ persons. If the A4IS is retained for any period and in whatever form, Center
' personnel should be informed thoroughly of its accuracy requirements, its
operation, capabilities, limitations, and basic purposes, perhaps through a
~ "short course" as suggested by TSSG and PPBS/PPS.
~ A reasonable approach to design of a management inforration systc~n (what
~ it should do and why, what it can do, and what it cannot do) is to concentrate
s upon the decision process, identifying the decisions to be made and the informa-
Lion needed to make them. These decisions are both planning and control decisions,
all of which vary in a ninnber of parameters. The material in Appendix E will
' not be repeated here, but it is obvious that, to a very great extent, the MIS
~ is conceived of as a gigantic data bank able to spew out specific answers on
immediate demand to verities of questions in usable form merely by pressing a
button. For example, such capabilities~as data for use in answering questions
(unspecified), compiling papers and reports, yearly surr~nations, etc., are called
for. This completely misses the point, although a_n fairness decision information
~~~; is implicit in much of what was asked for. I�at it is also obvious that merely
~ asking managers for answers as to what is needed is not nearly a precise enough
7
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
tii't)RKING PAPER � S-E-C-R-E-T ~ F~~RKINr PAPER
Main Report
manner in which to detexminc management information needs. 1Vhat must be dons is
to study the decision process in depth to identify the decision points, the Lypes
of decisions made, the frequency of these decisions, the relevant time frames
in which they must be made, and the data and inforrriation needed. Then the
system can be designed, with full consideration of both hardcvare and cost/
effective questions, to meet as many of the requirements as possible and to
incorporate legitimate external information requirements.
~'dhile the MTS is now adjudged inadequate in a nuJnber of respects, if alI
the inforniation cited in Appendix E, along with the new needs that will crop tzp
in the course of a redesign, were to be incorporated in a ne�~r system, we would
be approximately back where we started, and rest assured that any system pre-
dicated upon such an approach would be a failure, in addition to being excessively
expensive. If a new system is desired, an unhurried, in-depth study must be
undertaken to determine what information and data are actually needed by
management and a system designed to provide these. Thexe is no- o~Tizr viable
alternative to the successful design and institution of a worthwhile, new
system.
' This also implies taking a good look at what the Center is actually doing
and what it will be doing in the future: that is, definition of current and
projected outputs. (Some work has been accomplished in this direction recently
which could be of use.) This also concerns inputs and is essential because
it is the ultimate concern of management; that is, the process about which
decisions must be made--tivhat does and will the Center produce? Output is not
necessarily a visible product tivhich can be precisely measured, but explicit
recognition of and understanding about Coaster output must ba achieved. This
has been a problem in the past, and not only with respect to the biIS. For
' example, first-phase mission exploitation means many different things to
different people. To some it means the OAK, but obviously the hard copy OAK
i is only a part of the first-phase because other tangible and intangible products
are forthcoming from the exercise; e.g., briefing boards, cables, information
transfer from film to PI to analyst perhaps via telephone, enhancement of ,the
i PI's knowledge and expertise later reflected in a basic report or special study,
i etc. It should be clear that the hard copy OAK (i.t is possible that there may
' not even be such a report in the future) is not the only or even necessarily
~ the best measure of first-phase processing, that it does not adequately define
the output. Porhaps the Center should be viewed as an information processing
and transfer link in a cormnunications network with an output susceptible to some
means of information flow measurement, rather than trying to assess output
according to discrete, tangible products which really are only the media or
vehicles for the information transfer. This does not mean that information
on tangible output items may not be useful, but only that perhaps the emphasis
~, is misplaced, jVhat then are we really speaking about as initial output when a
8
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
� �
}1'OR.KING PAPER
Main Report
S-E-C-R-E-T
R10R}:ING PAPER
mission is received for exploitation? }ghat is the most significant attribute
of this process, and what should and can a management information system tell
us about the process? This is the type of analysis required.
The PPB Staff was recently asked to give OAK unit costs; these cannot be
derived objectively from the A4IS for the reasons alluded to above. If hard
copy OAK production were stopped tomorrow, but something analogous to first-
phase processing continued, it is conceivable that there could be relatively
insignificant savings, particularly in PI time, the point being that we do
not have the information needed readily available in the MIS to make decisions
in this axea with any degree of assurance,,nor do we know if it is possible to
obtain sue}i information in any quantifiable sense because the activity has not
been explicitly and adequately defined. 1'he gross figures available could be
of some use, along with a healthy input of judgment, in assessing the problem,
particularly if the question were directed tolvard the broad subject of all
first-phase processing, but beyond that there are serious difficulties. Nor
should be attempt to impute accuracy because we have some figures; attaching
a number to something does not make it a fact!
In spite of the many problems attendant to the current N!IS and its
utilization, the System does have positive aspects which should not be overlooked
or minvnized. The NSIS is used: to determine grossly how the Center uses its
manpower, how and on what the components spend their time, and the costs of
specific projects or categories of projects; for various planning purposes, and
in the documentation for the PPB System. As was stated, it is mainly a historical
record on which few current operating decisions are based; however, instances
of decisions based primarily on MIS data can be cited; for example', by IAS.
As PSG/RU pointed out, all of the D}IS data can obviously be put to same use,
although thexc is little evidence of a great deal of effort having een ex-
pended in this direction.
Perhaps the best way in which to look at the positive aspects of the A4IS
is to consider what the state of affairs would be if the MIS were suddenly
discontinued. Obviously, Center management needs some data similar to that
available in the R4IS; in fact, t}ie real problem is that more and also certain
different types of information are needed at~d some of. it in a more timely
fashion. If the N1IS were deleted., components would be forced to establish
systems, manual and/or automated (in addition to those systems already existing
outside the NIIS), to fill the gap, and at the top management level at least,
there would be a major problem of mutual reconciliation and interface of these
systems. There is absolutely no assurance that this would be any less costly
than retaining the current MIS.
9
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
jt,~RICING PAPER � 5-E-C-R-E-T - {4~RKING PAP1rR
Main Report
It must be remembered also that NPIC now has an ongoing, tested system
whose capabilities--and deiciences--are kno~,m quantities; to discontinue the
biIS means moving into an area of tivhich little is known and in which there undoubted-
ly would be many problems. Emplo~Tees are now familiar with the A~iIS, and its
particular demands upon them, and are used to the routine. Discontinuing the
MIS would mean discarding the positive aspects, and any systems replacing the
AiIS will have real costs relative to indoctrination and re-establi~:i~nent of a
routine.
This leads to the conclusion that, until something better is a~ ~.lable,
the basic MIS should be retained--not necessarily in its current cr :titration--
unless further study proves this to be a bad conclusion; that is, ~.f~: ~ s the
valid information requirements now met by the biIS can be met througt~..:Lher
decentralized and more efficient systems, probably basically manual ~~,Zd unit
record or ADP, whose operations and costs present a more favorable �e~~ernative.
However, before this can be determined, any and all other systems m~i:,t be Vrell
defined in further study so they can be weighed against the MIS. Iaa other
words, it would be a mistake to annotance discontinuance of the MIS and generate
probably what would be a mad scramble to establish other systems; if it were
decided to discontinue thelviIS and not replace it with a more powerful centralized
and automated system, the steps following the demise and any other systems to
be operated in lieu of the A~1IS should be well planned, understood, and agreed to
by a1T. There are positive improvements which could be made in the MIS, along
with certain decrease in costs to be achieved, particularly with respect to
manpower savings associated with data input and interpretation and manipulation
of output. In effect, the System can be trimmed and tailored to better meet
needs more efficiently, and perhaps its primary orientation redirected so that
its potential can be better realized while, at the same time, not discarding the
information useful to management; that is, the management information "fallout"
from the System.
A new system, based upon in-depth study of Center input and output, the
decision processes, and external i.nforniation requirements, would be focused
upon the management control area. However, the system should have the capability
of drawing upon data stored and used for operational control and strategic.
planning purposes. The application would therefore be an integrated data pro-
cessing system incorporating, firstly, pertinent human resource, materiel resource,
and financial information used for management decision-making and external
reporting, and, secondly, detailed operational control data, deemed necessary
and feasible to include in the system, and certain strategic planning informa-
tion. As pointed out in Appendix D, however, it is neither desirable nor
feasible for the system to support routinely the strategic planning activity.
10
S-E-C-R-E-T
Approved For Release 2008/09/23
iURKI~~G }~AI~ER � S-E-C-R-E-T
A4ain Report
WORKING PAPER
Therefore, flexibility is of paramount importance; that is, the capability
to add, delete, and alter both application programs and storage files, because
management information needs, much less operational control systems, are not
static over time and lead times are often such that a system can be outdated
' ~ before it is operational. This approach is now within the state-of-the-art.
Far example, with such a system, it would be possible to add new types of infor-
mation as needs change, to delete files no longer worthwhile to retain, and to
change programs W}tlc}1 assemble and manipulate data dra{vri fxom storage files.
This capability, possibly along {vitll real-time access and remote display devices,
{vould be built into and be an integral part of the system, obviating the exten-
sive and expensive reprogramming now attendant to any major changes to the A~IIS.
' Such a system is usually termed a computer-based, general purpose, a.nformation
management system; private industry is now working ort the concept.
Viable alternatives to the management i.rforn{ation problem are:
' 1. Discontinue the present A4IS; reject the centralized,
automated system concept, and establish the appropriate de-
centralized systems.
2. Discontinue the present MIS; use other interim systems,
and move toward a new, more po{verfirl, centralized, automated
system.
3. Continue the present IviTS, vnproving it as much as
possible, and do not consider other systeris.
4. Continue the present MIS, improving i.t as much as
possible, and move to{vard a ne{v, more powerful, centralized,
automated system.
Major factors, developed in the preceding discussion, bearing on the problem
are:
1. As a management information system, the current MIS
is conceptually invalid primarily because of its project
account orientation and its narro{v scope.
2. The MIS is now actually inadec}uatc as a management
information system, and for other purposes for that matter,
in a riumbex of respects, including timeliness, accuracy, output
formats, and data content.
3. The full potential of the current MIS }{as not been
realized.
11
CIA-RDP73T00325R000100010003-9
s
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
� ti~'CRICING PALER � S-}--C-}:-l;-T � tiy'ORKING PA}?Elt
Blain Report
4. The current MIS is not excessively costly in view
of its state-~d purpose, but it is costly in view of its actual
perfonnaiice.
5. The eurrertt AIIS can be modified to make it more efficient.
6. If the MIS is discontinued certain benefits will be
foregone.
7. There is no assurance that there will be any savings
to accrue from discontinuance of the MIS.
8. Replacement of the MIS with decentralized systems is
liable to be fraught with problems wherein significant information
gaps could emerge.
9. It is doubtful that other than a centralized, automated
system can meet the needs of Center management.
10. Except for one Group, there is unanimous interest in
more timely information, particularly in real-time capability
and remote output devices.
11. In certain cases, NPIC management conceives of a new
system as a gigantic data bank able to respond rapidly to almost
any question. This is not a. practical concept.
12. There are two fundamental objectives for a management
i information system: to provide essential information needed
� for management decision-making and to provide information
,f legitimately demanded by external authority. Any new system
1 should be designed to achieve only these objectives. Extraneous
~ and "nice-to-have" information should. be excluded from the
system.
i
13. T}ie type of system able to achieve the objectives is
Wlth1J1 the state-of-the-art.
Matching the alternatives tivith the major factors bearing on the problem,
the following conclusions and reco~runendations have been reached:
1. The Center should stick. with the concept of a
� centralized, automated management information system; to
reject this concept is a step backward. This is in line
,.-
12
S-E-C-R-E-T
i
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9
li'ORhIiv~ P~'ER � 5-1-C-R-Iu-T � ~VORKING Pt'1I'ER
Main Report
with the Executive Director�s stated desire that the Center
move toward automation and optimization of the exploitation
process.
2. The current AiIS should be retained and modified at
a reasonable level-of-effort.
3. The modified MIS should be kept in operation while
a study of management information needs, with emphasis on
t}ie decision processes and external information requirements,
is performed.
4. As a result of the study, a new system concept should
be developed.
5. Preliminary cost estimates for modification of the
current MIS, if feasible, and new systems capable of fulfilling
the needs stated in the systems concept should be generated.
6. Cost/effectiveness evaluation of. the alternatives
should be carried out and the results, along with the final
systems concepts, presented for executive decision.
7. The decision reached in step 6 should be pursued
vigorously and strongly backed by NPIC management through
to and after implementation. It will be not only desirable,
but necessary, to assign some of the best assets of the
Center to this task.
i
.j
i
8. Because development of a new systems concept will
require an in-depth study of NPIC with respect to current
and future input, output, decision processes, and external
information requirements, there will be worthwhile side
benefits from the study with applicability to other Center
problems.
9. Such a study should be run primarily in-House,
with contractual assistance as needed.
13
S-E-C-R-E-T
Approved For Release 2008/09/23 CIA-RDP73T00325R000100010003-9