TM 19A_Philco_2000_PERT_System_Feb1963 19A Philco 2000 PERT System Feb1963
User Manual: Pdf TM-19A_Philco_2000_PERT_System_Feb1963
Open the PDF directly: View PDF 
.
Page Count: 47
| Download | |
| Open PDF In Browser | View PDF |
COMP T
DIVISI
ELECTRONIC
AAP OCESSI G
SYSTEMS
R
PHILeo 2000
PERT SYSTEM
February 1963
PHILCO CORPORATION
A SUBSIDIARY OF
~Aotor?!?~jzan~
Computer Division. 3900 Welsh Road
Willow Grove, Pennsylvania
TM-19A
@
Copyright 1963, Philco Corporation
PREFACE
The PERT Manual is a reference manual for the Philco 2000
PERT System; it is intended for persons having a general
knowledge of PERT. Chapter I is a general review of PERT
terms as they are used in this manual. The remainder of the
manual describes specifically how to submit and run PERT
problems on the Philco 2000.
This manual (TM-19A) incorporates all changes announced for
the Philco 2000 PERT Manual TM-19, dated April 1962.
iii
CONTENTS
Page
Chapter
I
II
ill
IV
PREFACE •••••••••••••••••••••••••••••••••
iii
INTRODUCTION •••••••••••••••••••••••••••••
vii
GENERAL DESCRIPTION ••••••••••••••
PERT Terms •••••
• •••••••
Graphical Description of a Project
Establishing a Network •••••
Dummy Events and Activities.
Beginning and End Events
INITIAL INPUT FORMATS
Activity Cards
Control Cards •••
Project Card
Ending Card
Sample Input. . . . • . . . . . •
OUTPUT FORMATS.
Sample Output . .
DIAGNOSTIC AND ERROR CORRECTING ROUTINES
Introduction • • • • • • • •
Bad Card ••••••••••
Illegal Time Estimate
Card Code Error •••••
Event Table Packed •••
Activity Table Packed
Illegal Date
.. .
Activity Not Found •••
Blank Identifier ••••••
Concurrent Activities
Open End ••••••••
Missing Completion Date
•••••
Completion Date Entered Twice
Loop Error •••••••••
Control Card Error •••••••
Control Card Intercepted
Project Not on History Tape •••••
Example of Error Procedure • .
,
.
1
1
1
2
3
4
5
5
6
6
8
8
9
10
11
11
11
12
12
12
13
13
14
14
14
15
15
16
16
17
17
18
18
v
CONTENTS (Cont'd)
Page
Chapter
v
SUCCESSIVE RUNS
The History Tape
Updating ••••••
Input Data for Successive Runs
Activity Cards
Control Cards
A New ID Card •••••
Error Correction
21
21
21
22
22
23
23
SERVICE ROUTINES
Introduction
PERTSERV Card Formats ••
PRINTHIST
PUNCHIST ••
DELETE
COMPRESS
COpy ADD
LIST ••••
WRTSENT
REWIND
TAPE-TO-CARD •
BININPUT
MESSAGE
HALT
25
25
25
INPUT PREPARATION AND OPERATING INSTRUCTIONS ••
Input Preparation
Program Halts
Output' Processing
31
31
31
31
A
TYPE-OUTS ••
33
B
PROJECT SIZE
35
C
ADAPTING PERT TO AN OPERATING SYSTEM ••••••••
37
VI
VII
24
26
26
26
27
27
28
28
28
29
29
29
30
Appendix
vi
INTRODUCTION
The Philco 2000 PERT System allows PERT computations to be
performed on the Philco 2000. The following are special features
of the Philco 2000 PERT System:
•
Handles projects consisting of up- to 7400 activities and
3700 events.
•
Operates at extremely high speeds. PERT projects of 1000.
activities have been run on the Philco 212 in 7 seconds,
including input/output time.
•
•
Permits the use of symbolic event names.
Allows activities to be submitted in random order; the
Philco 2000 PERT System resequences them and creates the
project network.
o
Provides the option of giving either one or three time
estimates for each event.
o
PERT input may be on a special PERT tape or on the general
operating system input tape.
•
Allows for buffered input if the special PERT input tape is
used.
o
Program data for any project is contained in memory, thus
eliminating the need for intermediate tapes.
•
Includes a complete set of integrated prognostic, diagnostic,
and service routines.
•
Incorporates automatic machine methods of testing PERT
networks for consistency and legitimacy of input data.
•
Computes variance to determine the probability of meeting
scheduled dates.
•
The system maintains a complete history tape which permits
subsequent runs, and easy project modification and updating.
•
Input for successive runs may be on punched cards or on the
history tape.
vii
viii
•
Allows changes to be made to initial program data through
the use of a new ID Card, without destroying the original data.
•
Does not require a knowledge of any computer or of programming logic.
I. GENERAL DESCRIPTION
PERT is a process by which, after designating the tasks comprising
a project, establishing the interrelations between the various tasks,
and specifying the amount of time required to complete each task,
the crucial tasks of the project may be determined. Any delay in
accomplishing these crucial tasks will delay the final completion
of the entire project.
PERT TERMS
GRAPHICAL
DESCRIPTION
OF A PROJECT
In describing PERT as prepared manually or as solved on the
Philco 2000, the following common PERT terms are used:
o
A project is defined as a network of activities and events.
o
Activities are time consuming
•
Events are the termini of activities. They designate either
specific accomplishments or points at which the programs
start. Activities are separated from each other by events
which are used as activity identifiers.
o
The duration of an activity is the time period required to
complete an activity successfully.
•
The critical path is the specific sequence of events which
comprises the most rigorous time constraint in the accomplishment of the end event. If any event on the critical path is
delayed beyond the expected date of accomplishment, the
final completion date of the entire project can be expected to
be delayed by the same amount of time.
•
The slack of an event is defined as the time interval by which
the completion of a certain event can be delayed without
delaying the final completion date of the end event.
el~ments;
tasks to be achieved.
A project plan can be represented by a flow diagram. Each
activity of the project is depicted by an arrow. The inter-connection
among the arrows shows the sequence and interrelationship among
the events. The nodes of the graph correspond to the events of
the project.
1
)--------II~
Ot-----~
5
Figure 1. Sample PERT Pro;ect Flow Diagram
EST ABLISHING
A NETWORK
The first step in constructing a project network is to list the
activities that must take place in order to accomplish the end
objective. The activities may be listed in random order and do
not have to be sequenced.
The next step is to define the interrelation among the activities.
Two events are then associated with each activity, a preceding and
a succeeding event. These two events identify the activity and are
called the predecessor and successor events of the activity. An
activity cannot start until the predecessor event has occurred.
Similarly, an event cannot occur until all the activities that lead
to it have been completed.
The last step is to estimate each activity's duration. Three time
estimates should be made for each activity - the optinzistic ,
most likely and·pessimistic time estimates. PERT then calculates
the duration, utilizing the formula:
where:
dij
=
calculated duration of activity i,j
a
= optimistic time
m
=
most likely time estimate
b
=
pessimistic time estimate
estimate
Philco PERT can operate with the three time estimates described
above, or with only one estimate, the most likely time"
2
DUMMY EVENTS
AND ACTIVITIES
In order to construct the network correctly, it is sometimes
necessary to add dummy events and activities. For instance,
if two activities A and B begin and terminate with the same
events i,j (see Figure 2), a dummy activity and a dummy event (k)
should be added to prevent ambiguity. This activity has a duration
zero.
A
~mY
Event
i
j
B
B
Figure 2. The Use of Dummy Events
It can also happen that activities P and Q are predecessors of R,
while P is at the same time a predecessor of S; S, however, is
independent of Q (see Figure 3). It is therefore necessary to posit
a dummy activity D between P and R. The duration of D is zero.
,
p
~
S
0>
I D(Dummy Activity)
Q
~
R
..
Figure 3. The Use of Dummy 'Activities
3
BEGINNING
AND END EVENTS
It is necessary to introduce a unique event which initiates the
project. This event is the only one which has no predecessor.
Similarly, a unique event should terminate the project. This is
the . only event that has no successor. If a project starts with
several events (or ends with more than one event), one particular
event should be selected as the initiating (or terminating) event,
and the rest should succeed (or precede) it, utilizing dummy
activities as connectors.
4
II. INITIAL INPUT FORMATS
The input data for the initial run consists of a list of the activities
of the project, each activity being identified by its predecessor
and successor events. For each activity there should be one or
three time estimates. If there is a scheduled date for the completion of an activity, this date can be submitted with that activity.
In addition to the information given with each activity, the project
starting date must be given; specifying a deadline date is optional.
ACTIVITY
CARDS
Activity cards may be submitted in any order and should be prepared in the following format:
Column
1
*
Content
Card code. Should be 1 to indicate an
initial run.
2-10
A nine-character predecessor event
identifier. The characters can be any
legitimate Philco 2000 characters,*
provided they are not all spaces.
11-19
A nine-character successor identifier, restricted as above.
20-23
Optimistic time estimate. **
24-27
Most likely time estimate. **
28-31
Pessimistic time estimate. **
Refer to Philco 2000 Code Combinations, TF 17.
** The time estimates are four digits each and are given in tenths
of a week. The decimal point is not punched. If only one time
estimate is given, it should be punched in columns 24-27.
For example, a time estimate of 2 weeks is represented as
0020.
5
000000000000000000
1 l l • 5 ,
0
00
1 I I " II 12 1J 14 15 " II " " 10 21 12 21 24 15 U
00001000000001000000001000000010
nn
1I. 3t J2 Ui ~tl1 11 11 • •
0
010
00001000000
q" ulu 44 4\ U 4141" !II1~1 U U ~ 'fa !II \1 !illS." 11 \) ... e ,,'" .... n 12 n lIh, .. n 11 JI.
11111111111111111111· 111·· 11111' 1"1111111
11111111111111111111111' 111'111111-1-'111111
I
I
I
I
I'
I
o 222222222222" 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2' 2 2" 212 2' 22222122222222122 22222122222222122 2' Z Z" ZIZ Z- Z Z Z
..
I
I
I
I
I
1
::: 333333 33 3333333 3 3 3 3 3 3 3 3 33333333 3 313 3 3 3 3 3 3 313 3 3 3 33313333 3 3 3 313 3 3 333331333333331333333
o
I
I
I
I
1
1
o 4444444444 4444444444444 44444444 441444444441444444441 44444441444444441444444441444444
ee
f
5555
55
55
555
5555555555
5555555: 555555 5:5 5 55
5:5
555
55:555555 5
5~
555555 S} 55555
1
I
I
I
I
1
666666666666666666666666 6 6 6666666616 66666 616 6 6 6 6 6 6 61& 6 6 6 6 6 6 616 6 6 6 6 6 6 616 I 6 I • & &III' 6 III
I
I
I
I
I
I
777777777 7777 777777777777777777717777777717 7 7 7 7 7 7 17777 77717777777717 7 7 7 7 7 7 717 7 7 7 7 7
I
I
1
I
I
I
8 8888888888888888888888888888888818888888 1888888881888888881888888' 818 8 8 8888 II' I II"
ID:~TI:fYN~~D
1
4
L
i
LOCAT'I')N
I
I
COMMAND
1
12 1 14 I~ I TfIT1§" lr21 11 13 4 ~
5
2
11
------1
I 42 .]44
:J
d
U
41
ADDUls
AND
d !lI
51 il
ij
51
:MAUS :
!tot
i Si 5j $I !d iii Ii
S
:
6
I
1
4
Figure 4. Sample Activity Carel
Predecessor and Successor Events: HIRE 3ENG and DESIGN
Time Estimates: One Week, One and One-Half Weeks, and Two Weeks
Scheduled Date: December 15, 1962
CONTROL CARDS
Project Card
A project card must head each project deck and is punched
as follows:
Column
1-7
PROJECT
8
I (Indicates initial run)
9-24
Project identification
25-30
Project's beginning date
33-38
Project's end date (optional)
41
6
Content
1, if one time estimate is given;
blank, if three time estimates are given.
49-53
CARDS, if the history is on cards;
blank, if the history is on the history tape.
55-80
Remarks
D01OOO_o 0~00DTI0
1 J I • I I J I I '1 II 11 IJ MIS"
III
-
- 0000000
n ..
11 M 21
nn
: 2 222 2 2 2 2 2 2 Z Z Z 2 Z
000
24 It ,.
11111111111111111111
na
a 0 0 D10 0 0 0 0 0 0 010 0 0 0 0 0 0 010 0000 D0 010 0 0 0 0 0 0 010 O' Da a 0 DID. a 0 DG
II • 31 Jl JJ
1111$ •
U ....
u ulu« 4$
I
o
o
e~
5555
•
"
u ./•• 1111 U . . . . )I1
,
.hl • "
• • • " II q
I
II" •
I
I
I
I
I
1
33333333333 II 313 3 3 3 3 3 3313333333313 333333313333333313333333313 3 3 3 3 3
I
44
II U )01
I
Z 2 Z Z Z Z 2 2 2 2 2 2 Z Z 2122 2 2 2 2 2 Z12 2 2 2 2 2 2 212 2 2 2 2 2 2 212 2 2 2 2 2 2 212 2 2 2 2 2 2 212 2 2 2 2 2
333 3 3333 333
444444444
c, ..... hl
I
1
.. 33333
,..
1111' 11111111111111111111111111111111111111111111111111111111
1
I
1
I
I
4 444444444444444 (444144444444144444444144444444144444444144444444144 4 444
~
555 5 5 5 5 5 5 5 5 5 5 5 5 555 5 5 5 5 555555:555555 55:55555555:55555555:55 5 5 5 5 5:5 5 5 5 5 5 S 5:5 5 5 5 5 5
~
II. 16 & & 6 6 6 & & 6 6 6 6 6 6 &6 6 &6 6 6666
I
1
J
1
I
1
66666166666666166&666661666666661666666 I 616 6& 6 I I I 61111111
I
I
1
I
I
I
77 777 7 7 7 7 7 777 7 7 7 7 1 7 1 7 7 7 7 7 1 11117 11111 7 11111111111 7 11111 7 11 7 111 7 1 7 1111111 7 7 7 7 7717 7 7 7 7 7
•
I
1.888.8 I 8 •••• 8 • 8 .8888888 8 8 8 888888818
ID:I~~l(:~:D
I
I
"
I
I
1
I
I
1
a 8 a 8 8 8 818 8 8 8 ••• 818 8 8 8 8 • 8 818 8 8 • 8 8 8 818 a 8 8 8 • 8 811 I I • II
(",,)MMAND'
'OCAJJO,",
,
111
I
41"
1
ADOIf~S
44'"
AND .:1114 ... "'5
s
:
:
•
I
7
~1T1I"lf1flfn"ll
4
Figure 5. Sample Project Carel, Showing:
Project Identification: BUILDMISSILE-B
Project's Beginning Date: October 7, 1961
Project's End Date: Not Given
Number of Time Estimates: Three
Location of Project's History: History Tope
o0 0 0 0 o.
0 0 0 0 0 0 0" 0 ,,0 0" 0 0 0 . 0 0 0 0 0 0 O. 01. 0 0 0 0 0 0 ala
I , J f J , J • • "II "" Ie IS 11 IF lilt .11 n n 24 a H 112121.'1 U ~
111
1111$." •• 41
1111111,,111111111::1111::11:.111111
1111
I
I
52222 2 2 2 2 2 2 2 2 2 2" 2 .2" 2:. 2:. 2 2 22222 2 2 22:2 22
~ 33333
~
8
333
a a a 0 a 0 010 0 0 0 0 0 010 0 0 0 0 0 0 010 0 0 0 a 0 0 DID DODO D
.111 U U ~ 55 51 5151131 .1112 U M. _kr • • • n J2 n Min. n •••
, 11,1"".,11,1,111111,1111111111111111111111
U 41143" cs .. n _ u
I
I
I
I
2222:22222222:22 22222;22 2 2 22 2 2:2 2 2 2 2 2 22:22222 2
333333333333., 3333333333:33333333:333333. 3:33333333;33333333:33333333;333333
44 4 4 44 4 4 4 4 4 4 4 4444 .. 4 44 4 44 4 4 4 4 4 4 4 4 4 414 44 4 44 4 414 4 4 4 4 44414. 44444414444 4 4 4 414 4 44 U 4 4144 4 4 44
I
I
I
I
I
I
I
I
I
1
~ 5555 .. 5 5 5 5 5555555 5 5 5 5 5 5 5 5 5.,5555555 515555555515555555515555555515555555 515 5 5 5 5 5551555555
i
.. II
l
l
6 6 6 6 6 6 , 6 6 & 6 6 6 & & 6 6 6 6 & 6 6 6 & 6, & & 6 & 1& & & 6 6 6 616 6 6 6 6 6 6 616 6 6 6 6 6 6 616 6 & & 6 & 6 &1& 6 6 & & 6 I 616 & 6 6 I I
I
1717111
I
I
I
I
I
717117171117711777177111711117711111771711111117 711 71111717 7 7117 7 7 71 71171717 7
I
I
I
I
I
I
a 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 88888 8 818 8 8 888 8 818 8 8 8 8 8 8 818 a 8 a 888 818 8 8 8 8 8 8 818 8 8 8 8 8 a III 8 • II •
IO:~~~l'N~~O
L
LOC"''',)N
COMMAND
~
I
2-
1
1
ADD.I~
AND .:MAUS
:
"
1
•
I
7
Figure 6. Sample Project Carel, Showing:
Pro;ect Identification: POLARIS SUB-S123
Pro;ect's Begi~nin9 Dote: May 19, 1961
Pro;ect's End Dote: June 1, 1962
Number of Time Estimates: One
Location of Project's History: Cords
7
Ending Card
The last card of each project should have the character 9 punched
in columns 1 and 41. The rest of the card is ignored. Note that the
last project of an entire deck should also be followed by an END
card (see page 31).
SAMPLE INPUT
110161
010262
000000000000
ISTART
POWERSUPPOOOOOOOOOOOO
lSTART
WIRING
000000000000
ISTART
CARDSA
000000000000
ISTART
CABLES
000000000000
IPOWERSUPPUNITTEST 000600100014
lLAYOUT
CABLES
000600100014
lSTART
LAYOUT
000000000000
lCA~DSA
CARDSB
004000600070
lCABLES
SPATESTEOOOOB00120024
CABLESATT000600100012
lLAYOUT
lWI~ING
UNITTEST 001000140020
lCABLES
UNITTEST 002800420050
lCARDSB
UNITTEST 001000140020
lSPATESTEOUNITTEST 002800400044
IPROG~AM
SYSTTEST 001000140018
lCABLESATTSYSTTEST 004000600080
lUNITTEST SYSTTEST 002000280042
ISYSTTEST INSTALL 002000300040
000000000000
000000000000
ISTA~T
CABLES
000000000000
IPOWE~SUPPUNITTEST 000600100014
lLAYOUT
CABLES
000600100014
ISTA~T
LAYOUT
000000000000
lCARDSA
CA~DSB
004000600070
lCABLES
SPATESTE0000800120024
lLAYOUT
CABLESATT000600100012
lWI~ING
UNITTEST 001000140020
lCABLES
UNITTEST 002800420050
lCARDSB
UNITTEST 001000140020
ISPATESTEOUNITTEST 002800400044
IPROGRAM SYSTTEST 001000140018
lCABLESATTSYSTTEST 004000600080
lUNITTEST SYSTTEST 002000280042
ISYSTTEST INSTALL 002000300040
9
9
P~O~ECTISPACET~ACKCAD
lSTA~T
P~OG~AM
000000000000
lSTA~T
PROG~AM
ISTA~T
POWE~SUPPOOOOOOOOOOOO
ISTA~T
WI~ING
ISTA~T
CA~DSA
Figure 7. A Complete Input Card Deck for PROJECT SPACETRACKCAD
(Output from this project is shown in Figure 8, poge 10.)
8
III
0
OUTPUT FORMATS
The output from a PERT program consfsts of a complete list of
all the activities of the project including the dummy activities.
Each activity is identified by the original predecessor and
successor events of the activity card. The printed output for each
activity is described below.
o
The expected date is the earliest possible completion date of
the activity. Calculation of this date is based on the beginning
date of the project, the durations of the activities and the interrelations among the preceding activities.
o
The late date is the latest possible date for completion of an
activity without delaying the completion of the end event.
o
The
and
The
The
is a
o
The schedule date is the date specified on the activity card
in columns 32-37..
o
The actual completion date is the date on which an activity
was actually completed. (See Updating, page 21 and Input
Data for Successive Runs, page 22.)
slack of an activity is the difference between the late date
the earliest expected date for completion of the activity.
activities with slack zero are the critical path activities.
slack is printed in weeks.. For example, a slack of 2.5
slack of two and a half weeks ..
o
The duration of an activity is the calculated value a+4m+b/6
if three time estimates were made .. (See Establishing a Network, page 2.) If only one time estimate is made, the duration
is equal to that time estimate.
o
The program also calculates the variance of the duration.
Utilizing the computed 'variance it is possible to estimate
the probability of actually meeting scheduled dates.
9
PERT calculates the variance using the formula
" _
V 1J -
(b-a)
2
6
where:
SAMPLE OUTPUT
Vij
= The
a
= Optimistic time
b
= Pessimistic time
variance of activity i,j
estimate
estimate
The standard output does not include the variance. The following
is an illustration of standard PERT output.
PERT SYSTEM
SPACETRACKCAD
ACTIVITY
COMPo DATE EXP. DATE
START
PROGRAM
11/01/61
START
POWERSUPP
11/01/61
START
WIRING
11/01161
START
11/01/61
CARDSA
START
LAYOUT
11/01161
START
CABLES
11/01/61
SYSTTEST
PROGRAM
11/11/61
POWERSUPP UNlTTEST
11/08/61
WIRING
UNITTEST
11/11/61
CARDSA
12/12/61
CARDSB
LAYOUT
11/08/61
CABLES
LAYOUT
CABLESATT
11/08/61
CABLES
SPATESTEO
11/17/61
CABLES
UNITTEST
12/07/61
CARDSB
UNITTEST
12/22/61
CABLESATT SYSTTEST
12/20/61
SPATESTEO UNITTEST
12/15/61
UNITTEST
SYSTTEST
1/11/62
SYSTTEST
INSTALL
2/01/62
PAGE 1
LATE DATE SCHED. DATE SLACK
12/02/61
11/15/61
11/12/61
10/02/61
10/09/61
10/16/61
12/12/61
11/22/61
11/22/61
11/12/61
10/16/61
10/31/61
10/25/61
11/22/61
11/22/61
12/12/61
11/22/61
12/12/61
1/02/62
4.4
1.9
1.5
- 4.3
- 3.3
- 2.3
4.4
1.9
-
4.3
-
4.3
Figure 8. Output from PROJECT SPACETRACKCAD
(Input for this proi.ect is shown in Figure 7, page 8.)
10
1.5
4.3
3.3
1.2
3.3
2.2
4.3
1.2
3.3
DURATION
.0
.0
.0
.0
.0
.0
1.4
1.0
1.4
5.8
1.0
1.0
1.3
4.1
1.4
6.0
3.9
2.9
3.0
IV. DIAGNOSTIC AND ERROR CORRECTING
ROUTINES
INTRODUCTION
The PERT diagnostic program is a monitoring and correcting
program which tests the input cards for legitimacy of the characters and consistency of the data. Processed data is analyzed
for system errors such as internal loops within a project,
concurrent activities, open ends, and other trouble spots.
It also prints out a complete description of the errors or possible
errors and attempts to correct them. The programmed corrections
are not always sufficient, but they enable continued processing and
calculations that result in an initial output which provides at
least a general indication of the status of the project.
These corrections also facilitate later updating and corrections.
(See Updating, page 21 and Error Correction, page 24.) As a
result, errors can be corrected simply by submitting correction
cards, rather than by searching for bad cards in a large unsequenced deck, replacing them, and then reprocessing all the data.
The error output medium is the High-Speed Printer unless
otherwise specified"
Descriptions of PERT error print-outs and diagnostic procedures
follow.
BAD CARD
Description
Columns 1-40 of the input card are not identical to columns 41-80,
indicating a bad card-to-tape transmission and/or a keypunch
error.
Print-out
BAD CARD followed by the contents of the card in error. All
80 columns are printed.
Format
BAD CARD
Procedure
The program processes the data in columns 1-40.
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
11
ILLEGAL TIME ESTIMATE
Description
A time estimate is illegal if any of its characters are not numeric
(space is an illegitimate character), or if the three time estimates
do not satisfy the inequality a ~ m :s b.
Print-out
ILLEGAL TIME ESTIMATE followed by the contents of the activity
card in error.
Format
ILLEGAL TIME ESTIMATE
Procedure
PERT assigns the value zero to the duration and the value. 01 to
the variance.
XXXXXXXXXXXXXXXXXXXXX
CARD CODE ERROR
Description
An illegitimate character is detected in column 1 of the activity
card.
Print-out
CODE followed by the contents of the card in error.
Format
CODE
Procedure
The program interprets the code as code 1 and continues the
processing.
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
EVENT TABLE PACKED
Descri pti on
The number of events of a project is larger than the maximum
allowable number of events (see Appendix B).
Print-out
EVENT TABLE PACKED followed by the contents of the card that
caused the error or by the activity identifiers only.
Format
EVENT TABLE PACKED
or
CONCURRENT ACTIVITIES
XXXXXXXXXXXXXXXXXXX
EVENT TABLE PACKED I J
Procedure
The event is ignored and the program completes the processing
of the input data but does not perform the analysis and scheduling.
The processed data is put on the history tape or on cards; adjustments should be made during successive runs. (See Chapter V,
Successive Runs, page 21.)
Remarks
The event table could be packed as a result of too many events in
the input data, in which case the contents of the card containing the
event that could not be entered into the event table is printed.
If the event table is packed due to the need to add a dummy event
(see Concurrent Activities, page 14), the identifying events of the
concurrent activities are printed.
12
ACTIVITY TABLE PACKED
Description
The number of activities is larger than the maximum allowable
number of activities (refer to Appendix B).
Print-out
ACTIVITY TABLE PACKED followed by the contents of the card or
the identifying events of the activity that could not be entered into
the activity table.
Formats
ACTIVITY TABLE PACKED
or
CONCURRENT ACTIVITIES
or
OPEN END
or
OPEN END
xxxxxxxxxxxxxxxxxxxxxx
ACTIVITY TABLE PACKED IJ
ACTIVITY TABLE PACKED HAS
NO PREDECESSOR
ACTIVITY TABLE PACKEDI
HAS NO SUCCESSOR
Procedure
The activity is ignored and the program completes the processing
'of the input data, but does not perform the analysis and scheduling.
The processed data is stored on the history tape or on cards.
Adjustments should be made during successive runs"
Remarks
The activity table could be packed as a result of too many activities
in the input data. In this case, the card which contains the activity
that could not be entered into the activity table is printed. If the
activity which cannot be entered into the activity table is a programmed dummy activity, this information will be printed" (Refer
to Formats above; Blank Identifier, page 14; and Concurrent
Activities, page 14.)
ILLEGAL DATE
Description
An illegal date is detected on an activity card" The following are
illegal dates:
•
A date with a non-numeric character (e"g., spaces)
o
A date that does not exist, such as 15/1/60
•
A date in a year previous to the year or to the beginning of
the project.
Print-out
ILLEGAL DATE followed by the contents of the card which contains
an illegal date"
Format
ILLEGAL DATE XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX
13
Procedure
The date is ignored" If the date error concerns the initial date of
the project during an initial run, PERT processes the input and
puts the data on the history tape or on cards, but does not proceed
with the analysis and scheduling. If the date error concerns the
initial date during a successive run, the program uses the initial
date used in the previous run, and ignores the new starting date.
ACTIVITY NOT FOUND
Description
An activity with a code different from 1 (new activity card) is not in
the activity table (see Successive Runs, page 21).
Print-out
ACTIVITY NOT FOUND followed by the contents of the activity
card.
Format
ACTIVITY NOT FOUND
Procedure
The activity card is ignored.
XXXXXXXXXXXXXXXXXXXXXXXXXX
BLANK IDENTIFIER
Description
An identifier event which is all spaces (9 spaces) is detected.
Print-out
BLANK IDENTIFIER followed by the contents of the card.
Format
BLANK IDENTIFIER
Procedure
The identifier is changed to ????????? and the program continues
processing.
XXXXXXXXXXXXXXXXXXXXXXXXXXXX
CONCURRENT ACTIVITIES
Description
Concurrent activities are activities with identical predecessor
and successor identifiers.
Print-out
CONCURRENT ACTIVITIES followed by the predecessor and
successor identifiers of the two activities which are formed by
dividing one of the concurrent activities (see Procedure). The
first activity printed has the same duration and variance as the
original undivided activitYo The second activity is an added dummy
activit Yo If a dummy activity cannot be added, a further explanation is printed.
Formats
CONCURRENT ACTIVITIES
or
CONCURRENT ACTIVITIES
or
CONCURRENT ACTIVITIES
14
I
DUMMYn
DUMMYn
J
EVENT TABLE PACKED I J
ACTIVITY TABLE PACKED I J
Procedure
One of the concurrent activities I, J is broken into two activities
I, DUMMYn and DUMMYn, J. I, DUMMYn has the same duration
and variance as the original activity I, J; DUMMYn, J is a dummy
activity with the duration zero and variance. 01.
The dummy event DUMMYn is added to the event table and the
dummy activity is added to the activity table. The addition of
events and activities may result in packing the activity table or
event table. Should that happen, the dummy activity and dummy
event are not added and an appropriate error indication is
printed.
Remarks
PERT sequences the dummy events added to the project. The first
added event is denoted by DUMMY1, the second by DUMMY2 and
the nth by DUMMYn. (Refer to Card Code Error, page 12; Event
Table Packed, page 12; and Error Correction, page 24.)
OPEN END
o escri pti on
An event without a successor event which is not the end event, or
an event without a predecessor event which is not the initial event,
indicates an open end.
Print-out
OPEN END followed by the identifiers of the added dummy
activity.
Formats
OPEN END
or
OPEN END
or
OPEN END
or
OPEN END
Procedure
B
I
I
E
ACTIVITY TABLE PACKED I HAS NO
PREDECESSOR
ACTIVITY TABLE PACKED I HAS NO
SUCCESSOR
The program adds a dummy activity B, I in the case of an event
I without a predecessor (B being the initial event). It adds a
dummy activity I, E in the case of an event without a successor .
(E is the. end event). The dummy activities have durations zero
and variances .01.
MISSING COMPLETION DATE
D escri ption
Upon the submission of the completion date of an activity i, j,
PERT detects that the completion date of some activity k, i terminating with i is missing (see Updating, page 21).
15
Print-out
MISSING COMPLETION DATE followed by the contents of the
completed activity card.
Format
MISSING COMPLETION DATE
Procedure
The date is stored in a date list, but the processing associated
with the completion date is not done.
Remarks
Adjustment has to be made during successive runs. The printed
message does not indicate which completion date is missing.
xxxxxxxxxxxxxxxxxxxxx
COMPLETION DATE ENTERED TWICE
Description
Upon the submission' of the completion date of an activity i, j,
PERT detects that the completion date of some other activity
terminating with j was entered twice. (See Updating, page 21 and
Error Correction, page 24.)
Print-out
COMPLETION DATE ENTERED TWICE followed by the contents
of the completed activity card.
Format
COMPLETION DATE ENTERED TWICE
Procedure
The. completion date is stored in a date list, but the processing
associated with the completion date of an acitivity is not done.
Adjustment has to be made during successive runs.
XXXXXXXXXXXXXXX
I
!
RemC'uks
The printed message does not indicate which completion date was
entered twice.
LOOP ERROR
Description
A loop is a set of n consecutive events such as ai, ai + 1 ••• an in
which ai precedes ai + 1 and an precedes ai. Figure 9 illustrates
a loop formed by events C, D and E. Any loop of events is considered by PERT to be an error.
Figure 9. Sample Loop
16
.
I
Print-out
A list of all elements of the loop and the elements attached to the
loop.
Type-out
LOOP ERROR will be typed on the Console Typewriter.
Printer Format
11 12 13 ••••••
Proced ure
A loop error is a serious error which the program cannot correct. When a loop error is detected, the data is put on the history
tape or on cards without any further processing.
CONTROL CARD ERROR
1. Description
Procedure
2. Description
Procedure
3. Description
Procedure
4. Description
Procedure
5. Description
Procedure
Remarks·
An illegal character is detected in column 8 of the PROJECT card.
The project is ignored and a search is made for the next control
card.
BININPUT card is not preceded by a PROJECT card.
The contents of the BININPUT card is typed and a search is made
for the next control card.
A character other than a blank or 1 is detected in column 41 of
a PROJECT card.
One time estimate is assumed and the project is processed.
An illegal tape unit is specified in columns 57 -58 of the PERTSERV
card, i.e., a unit other than 0-15.
The PERTSERV card is ignored and a search is made for the next
control card.
An illegal PERTSERV routine is specified in columns 17 -26.
The PERTSERV card is ignored and the next control card is
searched for.
All CONTROL CARD ERROR type-outs occur immediately following the card in error.
CONTROL CARD INTERCEPTED
Description
A PROJECT, PERTSERV or END card is detected among the
cards of another project while processing, i.e., before a card
with code 9 is found. (See Input Preparation and Operating
Instructions, Chapter VII, page 31; also see PERTSERV, page 25.)
17
Type-out
CONTROL CARD INTERCEPTED followed by the contents of the
detected control card is typed by the Console Typewriter.
Formats
PROJECT CARD INTERCEPTED
xxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxx
PERSERV CARD INTERCEPTED
END CARD INTERCEPTED
Procedure
Action is initiated according to which control card (PROJECT,
PERTSERV or END) is intercepted.
PROJECT NOT ON HISTORY TAPE
Description
A project searched for by PERT or PERTSERV is not found on
the history tape.
Type-out
PROJECT NOT FOUND followed by the contents of the control
card.
Form at
PROJECT NOT FOUND
Procedure
PERT will search for the next control card.
xxxxxxxxxxxxxxxxxxxxxx
EXAMPLE OF ERROR PROCEDURES
Description
Figures 10, 11 and 12 show an input deck containing errors, the
PERT print-out indicating these errors, and the PERT output
wi!h these errors corrected, respectively.
Ifl
IP
1/
I I IJ
I
I
I
Ofnor:p4fJOfJ50
1 .
IiI
!
i
POJ~~~lfFROR Tq.~'~T===-rrr_tO_.t_~?_-~~____~~==~~=-~__~____~__~1
I. ~
~-t;Q. ':t:-~:AOOIUS AND '(M"',U
CI)MMANO
~ ~ ~ ~ ~ ~ ; ~ ~ ,~~ ~ ~,~ ~ ~ ,; ~ ~ ~ ~ ~ ~~. ~! ~:: ~~, ~ ~ ~:~ ~ ~ ~ ~ ~ ~ ~:~ ~ ~ ~ ~ ~ ~ ~:~ ~ ~ ~ ~!~, !;,~ ~ ~ ~ ~ ~! ~:~ ~: ~ ~ ~ ~ ~:~ ~ ~ ~ ~ ~
1 1 1I~ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 , 1 ,1 1 1 1 1 111 1 1 1 1 1 1 111 1 1 1 1 1 1 111 1 1 1 1 1 1 111 1 1 1 1 1 1 111 1 1 1 1 1 1 111 1 1 1 1 1
I
I
I
I
I
I
32222222222222222 .222222222222 .2222:22222222:222222 22:2 2 2 2 2 2 2 2:2 2 2 2 2 2 2 2:22 2 2 2222:22 2 222
~
~ 33333, ,3333333,33,333 3111313113131i133 333 3 313 3 3 3 3 3 3313333 333313 3 3 3 3 3 33 13 33333 3 313 33 3 3 3 ~
e
~
4444444444444444444444444444444444:4444' 4 4 4:' 4 4 4 4 4 4 4:4 4 4 4 4" ,:, 4 4 4' 4' ,:, 4 4 4"
4 ,:, 4""
I
I
I
I
I
I
I
1
I
1
I
I
~ 5555.,555,555555,5555555555555555551555555551555555 5 515 5 5 555551555555551555555551555555
x
L
~
;::
is
l-
Z
66 .666666661. 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6,6666616666666616666666616666666616666666616666666 616 6 6 666
I
I
I
t i t
-
I 111111111111111111111111111111111111.)1111111111111 111 111111111111111111111111 111 11111
I
I
I
I
I
I
8&8 8 88 888 888 88 8 8 88888&88 88888 B8 8 8 818 8 88888818 8 8 8 88 8 818 8 8 8 8 8 8 818 8 8 8 8 8 8 818 8 8 8 8 8 8 ~I!!.!!ll
!o:(~~EYN~~D 1$
2'
l 11 'OCJoYION
COMMAND
11I111.1~ll11.:'1211,1.
1
6
2:
1 AODU;S AND
401141'144445",')115151
':MAUI :
5
:
_6
:.
1
~~~ffT1615l1(t.\_~"t")iTl;;~IT1'i"lO_
Figure 70. Sample PERT Input Deck Containing Errors
18
-
i-
This input deck contains the following PERT input errors:
BAD CA~D
BAD CARD
BAD CARD
BAD CARD
BAD CARD
ILLEGAL TIME ESTIMATE
ILLEGAL DATt::
CODE
ILLEGAL DATE
BLANK IDENTIFIER
CONCURRENT ACTIVITY
A
OPEN END
•
BAD CARDS. Punching for the right-hand side of the cards
was begun in column 40 rather than in column 41.
•
ILLEGAL rIME ESTIMATE. The optimistic time estimate for
activity AB is illegal - it contains a non-numeric character
(a space).
•
ILLEGAL DATE. The date of activity AB is illegal since it is
earlier than the beginning date of the project. The date of
activity BC specifies the thirteenth month, and is therefore
illegal.
•
CODE. Activity BC contains Code 2 which is illegal for an
initial run.
•
BLANK IDENTIFIER. The last activity has an identifier which
is all spaces (the predecessor event).
o
CONCURRENT ACTIVITIES. Two activities are labeled AC.
2B
IA
IA
I
9
IA
IA
2B
2B
I
C
C
C
C
B
B
C
C
C
DUMMY 1
A
PERT SYSTEM
ERROR TEST
001000100010130161
001000200030010162
002000300040120162
003000400050
0020003 004010161
0020003 004010161
001000100010130161
001000100010130161
003000400050
DUMMY
lc
2B
IA
IA
I
C
C
C
C
PAGE I
001000100010130161
001000200030010162
002000300040120162
003000400050
IA
IA
2B
2tl
I
B
B
C
C
C
0020003 004010161
0020003 004010161
001000100010130161
001000100010130161
003000400050
?7?7?7??7
Figure 77. PERT Error Print-Out of the Input Errors Shown in
Figure 70, page 79
PAGE 2
PERT SYSTEM
ERROR TEST
ACTIVITY
A
A
DUMMY
A
C
C
177171177
B
DUMMY
COMPo DATE EXP. DATE
i/O 1/62
1/01/62
1/15/62
1
1/22/62
1/29/62
1/08/62
1/15/62
111111711
B
A
1
C
C
LATE DATE
1/01/62
1/22/62
1/29/62
1/29/62
1/29/62
1/29/62
1/29/62
SCHED. DATE SLACK DURATION
.0
0
.0
3.0
1/01/62
2.0
2.0
3.0
12/01/62
1.0
4.0
0
1.0
3.0
.0
2.0
Figure 72. Final PERT Output with the above Input Errors Corrected
19
v.
SUCCESSIVE RUNS
THE HISTORY
TAPE
The PERT System maintains a record of projects either on
magnetic tape (referred to as the history tape) or on binary cards
(referred to as cards). Data for each project on the history tape
or caJ;ds consists of an identification block followed by four tables
in the order listed below.
When preparing a new history tape, ablockof Z's must be written
at the beginning of the tape. The last project on the history tape
is followed by a block of Z's.
Table
Content
Activity Table
All project activities including the predecessor and successor activity identifiers
and the duration and variance of each activity. In this table, the predecessor and successor events are identified by their
sequence numbers rather than by their
alphanumeric symbols.
Event Table
All events in the project including .the
alphanumeric name and the sequence number of each event and the number of incomplete activities terminating with the
event.
Completed
Events Table
The completion dates of all completed
events. This table is put on tape only during
successive runs; it is always put on cards.
~--------------~--------------------------~--~~.. ~-------~
Scheduled
Date Table
UPDATING
All scheduled and completion dates of the
project activities.
After an initial project run, adjustments in the project network
may necessitate the addition or deletion of activities, changes in
time estimates, and adjustments in the initial date or in the project's deadline. Any errors encountered by the diagnostic program
will also require corrections.
Once the project starts, periodic runs may be necessary since
every time an activity is completed, the date of its completion
must be submitted. The completion date of an activity should not
be entered more than once.
21
When the history tape is· maintained, the obsolete history is
deleted from the history tape and the new history is added, unless a NEW ID card follows the project card of the deck. (Refer
to A NEW ID Card, page 23.)
INPUT DATA FOR
SUCCESSIVE RUNS
When the history is maintained on cards, the order of cards for a
successive run is as follows:
The Project Card
A PERTSERV BININPUT Card
PERT Binary History Cards
Activity Cards
An Ending Card (Code 9)
When the history is on the history tape, the PERTSERV BININPUT
card and the binary history cards are omitted.
Activity Cards
The activity cards for successive runs have exactly the same
fields as those for an initiial run, i.e., Columns 2-10 and 11-19,
are the activity predecessor and successor events, respectively;
Columns 20-23, 24-27, and 28-31 are the three time estimates;
and Columns 32-37 contain the date (if any is submitted) of either
the actual completion date or the scheduled date. On cards for
successive runs, columns 41-80 must be a duplicate of columns
1-40.
The major difference between initial and successive activity
cards is in the card code (Column 1)0 The card code indicates the
type of changes to be made and may be any number between 1- 5.
Code 1: Indicates a new activity to be added to the program.
Code 2: Indicates a change in the time estimates. If, in addition
to a change in the time estimates, a new scheduled completion
date is required, it may be indicated in the date field of a Code 2
card. However, for change of date only, Code 3 must be used..
Code 3: Indicates a change in the scheduled date. The date field
is the only field processed on Code 3 cards. A legal date in the
date field will be interpreted as a new scheduled date to be inserted, while a zero in the date field will be interpreted as a
cancellation of the scheduled date of the activity.
Code 4: Indicates a completed activity. The date field is the only
field processed on Code 4 cards. The date will be stored by the
program as the completion date of the activity.
Code 5: Indicates a deleted activity. The activity will be deleted
from the activity table.
22
Control Cards
The deck of cards for a successive run, as for an intial run, should
be headed by a PROJECT card. The last card of the project should
be a code 9 card as in the initial run. A PROJECT card could be
followed by a new ID· card.
The deck of cards for a successive run, as for an initial run, should
be headed by a PROJECT card. The for.mat of the PROJECT card is
as follows.
Content
Column
1-7
PROJECT
8
S, indicating the history is on the history
tape
C, indicating the history is on cards.
A New 10 Card
9-24
The project ID - This should be the same as
that of the initial run of the project"
25-30
The initial date of the project.
33-38
Ending date" If this field is blank, there is
no change in the deadline. A legal date replaces the formerly used date; an illegal
date is ignored.
39-80
Ignored.
When making changes, it might be desirable to keep both the
changed and unchanged history on tape. This could, for example,
be desirable for comparison purposes. Should this be the case,
a NEW ill card must follow the PROJECT card. The updated
history will then be on the history tape under the new identification. The NEW ill card feature may be used only when the project's history is maintained on the history tape. An illustration
of a NEW ill card for a project is shown in Figure 13.
000000 D.
00000000000000000000001000000001 00000001
000000100000 a 0 010 0 a a 0 0 a 010 a a DOD
• 2 2 • I I I I I 11 " '2 II MIS. II . . . II 2' n D " "Jt JIll 1111 1I!lIl 11111 lIl1l1l1 .. II 12111 .. 11 II " " It _1M U 1111 • _ II _I_ . 1112 12 II • • iII •• II n n n lIllI • n II II •
111111111111111- 11-11111111111111111111111111111111 1111111-111-11111111111111111111111
00
I
I
I
I
I
I
: 2222222222 - - 2 2 2 2 222222222222222221222222221222222221- - 2 2 22 - 212 2 2 22222122222222122222 2 ~n
1
1
I
1
I
1
• 333333333 3 3 3:-: 3 3 3 3 3 3 3 33333333333333133333333133333333133" 3 3 3 3 313 3 3 3 3 3 3 313 3 3 3 3 3 3 313 3 3 333 ~
I
1
I
1
I
I
g
444 4 4 - 4 4 44444444444444 4 444444444414444444414 4 4 - 44 - 414 4 44 4 4 4 414 44 44 44 4144 4 44 H 41H 44 U ~
e~_ - 5 5 555555555 - 5 5 5f
5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5:5 5 5555 - ":55555555:555 - 5 55 -:555555 5 5:5 5 5 5 5 5 5 5:5 55 5 55
I
I
I
I
I'
I
::!
~
, 6 - 666666 6 6 6 6 6 6 6 6 666666666666666 6 616 6 6 666661- 6 &&6 6 6 61& 66666661666 I 6 6 6 118 8 8 III I 118 6 I I I I
. I
I
1
I
1
I
7 7 7 7 7 7 7 7 1 7 7 7 7 7 7 1 1 7 7 7 7 7 7 7 1 7 7 7 1 7 7 1 7 717 7 7 7 7 7 7 717 7 7 7 7 7 7 711 7 7 7 7 7 7 7.7 7 7 7 7 7 7 717 7 7 7 7 7 1 717 7 7 11 7
I
1
I
8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 I 8 I 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 818 8 8 8 8 8 8 818 8 8 8 8 8 8 81 a 8 8 8 8 I
IDS((N~~l;'~:D
L
~
LOCATION
COMMAND
1114"1111111]41&
1
2
I
1 AOO.;$
AND
I
1
I·
a 818 I 8 8 8 8 a 818 a I I I I I a18 I •• I I
.:MA.KS :
S
:
•
I
7
.1441444~!IISI52ijSi56Si5JSlSiCilI1
Figure 13. New 10 Card Sample
23
The NEW ID card format is:
Content
Column
1-7
NEWID
9-24
New project identification
41-47
NEW ID (same as 1-7)
49-64
New project identification (same as 9-24)
The rest of the card is ignored.
ERROR CORRECTION
During successive runs the errors of previous runs should be
corrected. If the errors occurred in the time estimates or date
fields only, a change in this field is sufficient. For example, if
the error was an illegal time estimate, a code 2 card should be
used to correct the error. If the error involves one of the
identifiers of an activity, two correction cards must be used: (1)
a code 5 card to delete the incorrect activity and (2) a code 1 card
to insert the corrected activity.
In case of concurrent activities, the programmed DUMMY activities must be replaced even if the program correction was
adequate. The new activity could be a dummy activity, but could
not have an identification DUMMYn, since in successive runs,
concurrent activities may recur and the program starts adding
the events DUMMY1, DUMMY2, " ••• , and ambiguity may occur.
24
'VI. SERVICE ROUTINES
INTRODUCTION
PERTSERV is a service program incorporated into the Philco
2000 PERT System. It provides the following service functions:
•
Prints out the complete data of any project on the history
tape.
•
Copies projects from one history tape onto another.
•
Adds or deletes projects.
o
Transmits the history of a project from the history tape to
binary or Hollerith punched cards, and from binary punched
cards to tape.
o
Analyzes proj ects for loops.
•
Lists all the projects on a history tape.
o
Initializes and rewinds history tapes.
Each reference to a service routine is made by a PERTSERV
control card. A PROJECT card may not immediately precede a
PERTSERV card except when it is a PERTSERV BININPUT
card followed by the binary history cards. PERSERV cards can
follow each other or any project's deck of cards, or may precede
any project decko
PERTSERV CARD
Column
Content
FORMATS
1-8
The control word PERTSERV.
17-32
The name of the service routine to be used.
33-48
The ID of the project, if an ID is necessary.
57-58
The tape unit in decimal, if a tape unit must
be specified.
Any columns not specified must be left blank.
The service routines available with PERT are described below.
25
PRINTHIST
Action
PRINTInST prints out the history of the project from the specified
history tape. This history .is a lif;)t of all activities with their
durations and scheduled and actual completion dates. The history
tape is positioned at the end of the printed project.
Card Format
Columns 1-8:
PERTSERV
Columns 17-25:
PRINTInST
Columns 33-48:
ID
Columns 57-58:
TAPE UNIT
If columns 57-58 are blank, tape 10 is assumed.
Example
PERTSERV
PRINTmST
BUILD AEROA1
10
PERTSERV will search tape unit 10 for the project BUILD
AEROAl. When the project is found, the history will be printed
out.
PUNCHIST
Action
PUNCInST punches the history of a project from the specified
history tape 10 in the input format (see page 5) with one time
estimate only.
Card Format
Columns 1-8:
PERTSERV
Columns 17-25:
PUNCmST
Columns 33-48:
ID
Columns 57-58:
TAPE UNIT
If columns 57-58 are blank, tape 10 is assumed.
Example
PERTSERV
PUNcmST
BUILD AEROA1
9
PERTSERV will search tape unit 9 for the project BUILD
AEROA1. When the project is found, a deck of cards of the project
will be prepared on output tapes. The cards should be punched
off line in code mode.
DELETE
Action
DELETE removes the designated project from the specified
history tape. The history tape is then positioned at the end of the
deleted proj ect.
Card Format
Columns 1-8:
PERTSERV
Columns 17-22:
DELETE
26
Columns 33-48:
ID
Columns 57-58:
TAPE UNIT
If columns 57-58 are blank, tape unit 10 is assumed.
Example
PERTSERV
DELETE
BUILD AIR01A
7
PERTSERV will search tape unit 7 for the project BUILD
AIR01A; when the project is found, it will be deleted from history tape 7.
Remarks
A project is deleted from a history tape by altering certain data
in the ID block. The remainder of the data is unchanged. If several
projects have been deleted from the history tape, it is desirable
to use COMPRESS.
COMPRESS
Action
COMPRESS copies all the non-deleted projects from tape 10 to
the assigned tape. The data of the deleted activities is ignored.
Both input and output tapes are rewound when the copying is
complete.
Card Format
Columns 1-8:
PERTSERV
Columns 17-24:
COMPRESS
Columns 57-58:
TAPE UNIT
Example
PERTSERV
COMPRESS
6
PERTSERV will copy all non-deleted projects from tape unit 10
to tape unit 6.
COpy ADD
Action
PERTSERV copies the specified project from the history tape on
unit 10 to the history tape on the specified tape unit. A sentinel
block of Z ' s is written following the copied project. Both input
and output tapes are positioned at the end of the copied project.
Card Format
Columns 1-8:
PERTSERV
Columns 17-20:
COpy
Columns 34-48:
ID
Columns 57-58:
TAPE UNIT
Example
PERTSERV
COpy
BUILD AIR01A
9
Project BUILD AIR01A will be copied from the history tape on
unit 10 onto the history tape on unit 9.
27
Remarks
When copying a project onto a history tape, PERTSERV looks for
a sentinel block and writes the specified project on top of the
sentinel block just beyond the last project on the tape. Therefore,
if there are no projects on the assigned tape, the COpy control
instruction should be preceded by a WRTSENT instruction (see
WRTSENT below).
LIST
Action
Types out a list of all projects on the history tape on the specified
tape unit. At the end of the list, END OF LIST is typed out. The
history tape is rewound at the end of the operation.
Card Format
Columns 1-8:
PERTSERV
Columns 17-20:
LIST
Columns 57-58:
TAPE UNIT
If columns 57-58 are left blank, tape unit 10 is assumed.
Example
PERTSERV
LIST
3
A list of all projects of the history tape on unit 3 will be typed out.
Remarks
It is advisable to follow COMPRESS, DELETE and COpy cards by
a LIST card.
WRTSENT
Action
PERTSERV writes a sentinel block on the specified tape unit. The
output tape is positioned at the beginning of the sentinel block.
Card Format
Columns 1-8:
PERTSERV
Columns 17-23:
WRTSENT
Columns 57-58:
TAPE UNIT
Example
PERTSERV
WRTSENT
11
A sentinel block will be written on tape 11.
REWIND
Action
PERTSERV rewinds the specified tape unit.
Card Format
Columns 1-8:
PERTSERV
Columns 17-22:
REWIND
Columns 57-58:
TAPE UNIT
Example
PERTSERV
REWIND
PERTSERV will rewind tape 10.
28
10
TAPE-TO-CARD
Action
Card Format
PERTSERV will 'put the history of the specified project from the
history tape mounted on a specified unit onto the normal output
tape. Binary cards could be punched off line from this tape. The
binary punched card format is as follows:
Columns 1-4:
The first four characters of the project's ID.
Columns 5-8:
Sequence numbers.
Columns 9-80:
The history in image mode.
Columns 1-8:
PERTSERV
Columns 17-26:
TAPETOCARD
Columns 33-48:
ID
Columns 57-58:
TAPE UNIT
If no tape is specified, tape 10 is assumed.
Example
PERTSERV
TAPETOCARD
BUILD AEROIA
9
Project BUILD AEROIA will be copied from tape 9 onto the output
tape for punching cards. Cards punched in the format described
above could be obtained from the tape using off-line tape to card
equipment.
BININPUT
Action.
PERSTSERV r.eads the binary history cards from the input tape
into memory and then transfers control to PERT for processing
and updating..
Card Format
Columns 1-8:
PERTSERV
Columns 17-24:
BININPUT
PERTSERV
BININPUT
Example
The project on the history tape will be read into memory..
MESSAGE
Action
PERTSERV types out the message written in columns 33-80 of the
PERTSERV card.
Card Format
Columns 1-8:
PERTSERV
Columns 17-23:
MESSAGE
Columns 33-80:
A message to be typed
29
Example
PERTSERV
MESSAGE
MOUNT TAPE 242 ON 9
PERTSERV
MESSAGE
PUNCH CARDS FROM TAPE
6
HALT
Action
PERTSERV types out the message written in columns 33-80 of the
PERTSERV card, and then halts. PERT proceeds to the next
project or PERTSERV card as soon as the ADVANCE bar is
pressed.
Card Format
Columns 1-8:
PERTSERV
Columns 17-23:
HALT
Columns 33-80:
A message to be typed
Example
PERTSERV
HALT
MOUNT SCRATCH TAPE ON 6
VII. INPUT PREPARATION AND OPERATING
INSTRUCTIONS
INPUT
PREPARATION
A number of projects may be processed at the same time. The
project decks should follow each other; PERTSERV cards can be
placed anywhere ahead of or between projects. (Refer to PERTSERV, page 25.)
AN END card should terminate the entire deck. An END card has
the word END in columns 1-3; the rest of the card is blank.
Input cards should be transmitted off line in code mode, ten words
per card, twelve cards per block.
The input tape should be placed on tape unit
o.
The history tape should be placed on tape unit 10.
PROGRAM HALTS
PERTSERV Halt
Instruction
PERTSERV control instruction HALT will type out a message
and then halt (see HALT, page 30). After obeying the instruction,
the operator should press the ADVANCE bar and the program will
continue.
Irrecoverable
In the event of irrecoverable tape error, the program types out:
Tape Error
10 TROUBLE
and halts o
At this point the operator should press the ADVANCE bar. PERT
will then attempt to process the next projecto
Final Halt
When all projects have been completed, PERT types out:
END PERT
and halts.
OUTPUT PROCESSING
All output is written on tape unit 5. In order to obtain both the
printed data and the punched cards, the procedure below should
be followed:
o.
•
For printed output, use Data Select
o
For binary punched cards, use Data Select 1.
o
For Hollerith punched cards, use Data Select 2.
31
The punched card decks are separated as follows:
•
The first card of each binary card deck has punches in all
rows of the first eight columns.
•
Each Hollerith deck is followed by several blank cards.
The last block of each project deck contains a conditional
stop.
The Hollerith punched cards should be interpreted.
32
APPENDIX A
CONSOLE TYPEWRITER TYPE-OUTS
The system starts with the type-out:
PERT SYSTEM INITIALIZED
For each project, the new ID card (if such a card follows the
project card) and the number of blocks on the history tape are
typed-out. All PERTSERV cards are also typed out. When all
projects have been processed, PERT types:
END PERT
PERT SYSTEM INITIALIZED
PROJECT CONTRUCTION S123
HISTORY 20 BLOCKS
PROJECT MFG SN2
NEW ID MFG SN3
HISTORY 6 BLOCKS
PERTSERV
PRINTHIST
AEROA1
6
PROJECT ID
mSTORY N BLOCKS
PERTSERV
MESSAGE
TAPE 242 ON 10
END PERT
33
In the event of an irrecoverable tape error the program types out:
10
TROUB~E
and halts. If the ADVANCE bar is pressed, the program proceeds
to the next project.
Notes:
The error type-outs are described in Chapter IV, Diagnostic
and Error Correcting Routines, pages 11-19. PERTSERVtype-outs
are described in Chapter VI, Service Routines, pages 25-31.
34
APPENDIX B
PRO JECT SIZE
The maximum size of a PERT project is a function of the size
of memory of the Philco 2000 system used and of the specific
operating system. The number of activities, A, is:
A
__ M - S - 4330
---:::--::::-::----- - 46
3.75
where M is the size of memory, and S is the size of the operating
system" The maximum number of events is always half the maximum number of activities.
For a 32,768 word Philco 2000 computer operating under SYS,
A = 32,768 - 512 - 4330 - 46
3.75
yielding a maximum PERT project of 7400 activities"
35
APPENDIX C
ADAPTING PERT TO AN OPERATING SYSTEM
A tape table is incorporated in the PERT System which allows
easy adaptation of PERT to operating systems other than SYS. The
table consists of 16 consecutive locations and is assigned the symbol TPUNIT by the card:
•
ASGN
TPUNIT,TAPETBL
The input, output and history tapes are assigned the symbols
UNITOX, UNIT5X, and UNIT10X, respectively, by the cards:
ASGN
UNITOX, TPUNIT +s
ASGN
UNIT5X,TPUNIT+t
ASGN
UNIT10X, TPUNIT+u
where s, t, and u are relative positions within the table.
Logical tape unit numbers are at T23 ·of table locations. References to tape units are made indirectly. For example, if columns
57-58 of a PERTSERV card contain 12, the actual tape unit
referenced is in location TPUNIT+12 at T23"
Data select reassignments can be made at compile time by
changing the following cards:
ASGN
PRINTSL,x
ASGN
DATAPUB,y
ASGN
DATAPUH,z
where x, y, and z are integers from 0-15 and are data selects for
printing, punching binary cards, and punching Hollerith cards,
respectively.
37
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : No XMP Toolkit : Adobe XMP Core 4.2.1-c041 52.342996, 2008/05/07-21:37:19 Create Date : 2018:05:15 16:32:49-08:00 Modify Date : 2018:05:15 16:34:37-07:00 Metadata Date : 2018:05:15 16:34:37-07:00 Producer : Adobe Acrobat 9.0 Paper Capture Plug-in Format : application/pdf Document ID : uuid:f715d855-0915-cc46-924f-6ce75edffd4a Instance ID : uuid:5f49e97d-f4df-ae40-9fc9-235715b9afe5 Page Layout : SinglePage Page Mode : UseNone Page Count : 47EXIF Metadata provided by EXIF.tools