See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/339438626

Linear Programming based Effective Maintenance and Manpower Planning

Strategy: A Case Study

Article · January 2008

CITATIONS READS

14 2,028

2 authors, including:

Buliaminu Kareem
Federal University of Technology, Akure
187 PUBLICATIONS 406 CITATIONS

SEE PROFILE

All content following this page was uploaded by Buliaminu Kareem on 23 February 2020.

The user has requested enhancement of the downloaded file.

B. Kareem and A.A. Aderoba

Linear Programming based Effective Maintenance and

Manpower Planning Strategy: A Case Study

B. Kareem and A.A. Aderoba

Department of Mechanical Engineering

Federal University of Technology, Akure, Nigeria
E-mail: [email protected]

Abstract organization resources can be resolved by

applying linear programming techniques.
In this study, maintenance related data Linear programming is a widely used
of a cocoa processing industry in Akure, mathematical technique designed to help in
Ondo State of Nigeria were collected, planning and decision making relative to the
classified and analysed statistically. Linear trade-off necessary to allocate resources
Programming (LP) model was formulated (Joseph, 1987). In any production firm there
based on the outcomes of the analysed data. are two sub systems, human and technical.
The data analysed includes maintenance The two sub-systems must be balanced and
budget, maintenance cycle, production coordinated in order to function effectively
capacity and waiting time of production (Joseph, 1987). Many studies have been
facilities in case of failure. Data were carried out on how to make maintenance and
analysed based on manpower cost, machine manpower planning effective in a production
depreciation cost and the spare part cost, firm. In previous studies, models adopted to
which were assumed to be proportion to the analyze prevailing situation include:
number/magnitude of the breakdowns. The simulation model, queue model, utility
generated LP model was solved using model and network analysis. In this study,
software named “the Quantitative System for linear programming technique is used to
Business- QSB (Version 3.0). The results of analyze maintenance operations and
the model showed that four maintenance manpower planning in a production firm
crews were needed to effectively carryout used to analyze maintenance operations and
maintenance jobs in the industry. The manpower planning in a production firm that
sensitivity analysis showed that the results was used as a case study.
have a wide range of feasibility.
Allan et-al (1988) defined maintenance
Keywords: Maintenance System, Manpower as all activity in keeping equipment in
Planning, LP-model, QSB Software, working order. They also pointed out that the
Sensitivity Analysis. amount and type of maintenance applied
depends strongly on its cost and safety
implication of the system failure. Joseph
1. Introduction (1987) said that both passive and active
repair times are influenced by factors other
Many operation management decisions than equipment design. He said maintenance
involving making most effective use of philosophy plays an important role in

26
 Linear Programming based Effective Maintenance and Manpower Planning Strategy: A Case Study

determining overall availability, and defined while the term programming in the context
it as consideration of maintenance means planning of activities. The objective is
procedures, personnel and spare part to maximize or minimize a single objective
provision. function relative to a set of constraints.

Armstrong (1971) said that manpower A mathematical program is linear, if

refers to human resources used in carrying F (x1, x2,…,xn) and each
out jobs in any organization. Report of the G (x1, x2,…,xn) (i =2,3,…,n)
conference board on manpower planning and are linear in each of their argument.
evolving system in 1971 further defined
manpower planning as a process intended to That is,
assure an organization that it will have the F(x1,x2,…,xn) = c1x1+c2x2+…+cnxn
proper number of properly qualified and and
motivated employees in its workforce at g (x1,x2,…,xn) = a11x1+a12x2+…+dinxn
some specific future time to carry on the ci and aij (i = 1,2,…..n; j = 1,2,…,n)
work that would then have to be done. Many are known constants.
scientific approaches have been used to solve
the problem of manpower. Among these are All linear programming problems have
the workload method, mathematical the following properties in common: all seek
programming approach and simulation to optimize some quantity. This property is
techniques. In all these approaches the main referred to as the objective function. There
aim of manpower planning that is pursued are constraints, which limit the degree to
are to ensure that the organization: obtains which the objective can be pursued; and
and retains the quantity and quality of there must be alternatives to choose from.
manpower it needs; makes the best use of its The objectives and constraints in linear
manpower resources; and is able to programming must be expressed in terms of
anticipate the problems arising from linear equation or inequalities
potential surplus or deficit of manpower. In
this study linear programming approach is However, the following procedures are
used to determine optimal number manpower necessary when formulating Linear
that can work effective in maintenance sector Programming (LP) problems: write down
subject to manpower cost, spare part cost and decisions variables of the problem; formulate
depreciation cost of machine. Many of the the objective function in terms of decision
existing models on manpower planning variables; formulate the other
treated the above constraints separately. The conditions/constraints of the problem to
there are scanty work on model that which the optimization process is subjected
holistically considered the three constraints. to, such as resources limitation, market
constraints as linear equations in terms of the
Overview of Linear Programming Model variables; add non-negativity conditions/
constraints- the considerations that negative
Linear programming was developed in values of physical variable in most cases do
1947 by G.B. Dantzig and represents a not have any valid physical interpretation.
valuable method in the field of operations Summarily, the objective function, the set of
research. Linear programming is a constraints and the non-negativity together
mathematical tool for finding solution to a form the linear programming model of the
certain class of problem. The word (linear) problem.
implies that the relations involved are linear,

International Journal of The Computer, the Internet and Management Vol.16. N.o.2 (May-August, 2008) pp 26 -34

27
B. Kareem and A.A. Aderoba

2. Methodology Adopted The budget for the production section of the

firm of which the maintenance section is a
In pursuing the objective of the study, subsection for the year is as shown in Table
data were collected, classified, analysed and 1a. Table 1b shows maintenance cycle,
the linear programming model was production capacity and average waiting
formulated based on the data analysed. The time of each machine.
department whose data was studied is the
maintenance section of a production firm In the analysis of the data collected, the
(Cocoa Processing Industry) in Akure, Ondo following assumptions were made:
State of Nigeria, which is responsible for the manpower cost associated with the
keeping of the plant and machinery used for maintenance of each machine per hour is the
cocoa processing in operable condition. The same for all the machines; depreciation cost
data were collected through the use of associated with each machine is directly
questionnaires and oral interview among proportional to the rate of breakdown of the
employees in the maintenance section of the machine; and spare part cost associated with
firm. The data that were collected include the the maintenance of each machine is directly
following: number and list of all the proportional to the number of hours used
machines; types of maintenance applied; during the maintenance of the machine.
budget on the maintenance; factors affecting Based on these assumptions, the data are
maintenance; present level of manpower analyzed as followed.
planning in maintenance department;
maintenance cycle of each of the machines; Total manpower cost for Maintenance
and the waiting time of each of the machines. Section = 19 x 6,900,000
51
After the collection of the data a close = N2,570,588.24./Year.
monitoring of the maintenance operations of
the production section was done over a Manpower cost associated with the
period of two weeks to make ensure maintenance of each machine was calculated
reliability of the data. The major machines using
on which scheduled or time based preventive D x 2,570,588.24
maintenance was carried out include Y 5760 per hour.
machine one (1) cleaning and destoner, two
(2) dryer, three (3) winnower, four (4) Where, D is waiting time of the machine
reactor, five (5) roaster, six (6) map mill, during a year and Y is the total waiting time
seven (7) liquor press, eight (8) butter press of all the machines in the production section
and nine (9) boiler. The factors affecting the in a year.
maintenance operation of the firm include
understaffing in the maintenance section, Spare part cost for each machine was
mismanagement of budgetary allocation, calculated using
inadequate tools, equipment and spare part.
D x 4,000,000
The total number of employees in the Y 5760 per hour.
maintenance section is nineteen,
maintenance manager inclusive. The Depreciation cost associated with each
employees are grouped into six crews, each machine was calculated using
of which has three members. Among the Q x 52,000,000
three members, one acts as the supervisor. W 5760 per hour.

28
 Linear Programming based Effective Maintenance and Manpower Planning Strategy: A Case Study

Table 1a: Budget for Maintenance The variables used are:

Operation X1 = Number of crew allocated to
machine 1
Manpower remuneration for the N 6,900,000:00 X2 = Number of crew allocated to
production section. machine 2
X3 = Number of crew allocated to
Total allocation for spare part N 4,000,000:00
machine 3
for maintenance of the machine.
.
Total cost for depreciation of N52,000,000:00 .
the machine. X9 = Number of crew allocated to
machine 9

Table 1b: Maintenance Cycle, Production The objective is to maximize the

Capacity and Waiting Time. percentage production hour available per
maintenance cycle of each machine. That is
M/C Maintenance Waiting Production minimize the waiting time of each machine.
Cycle, (Hrs) Time Capacity
(Codes) (H1) (Tons/hr) The objective function is written as:
1 720 (i) 1 3
Maximize Z = 0.998X1 + 0.998X2 +
2 1384 (ii) 2 2
0.999X3 + 0.998X4 + 0.997X5 + 0.994X6 +
3 5760 (iii) 2.5 2 0.996X7 + 0.999X8 + 0.998X9
4 1800 (iv) 3.33 2
5 2000 (v) 5 2.5 Subject to:
6 720 (vi) 6 1.5 Constraint on manpower cost associated
7 2000 (vii) 8 1.5 with the maintenance of each machine
8 1600 (viii) 1 2.5
13.49X1 + 15.74X2 + 5.62X3 + 17.90X4 +
9 2000 (ix) 3 1.5
28.1X5 + 53.96X6 + 44.97X7 + 6.75X8 +
16.86X9 ≤ 347.22.

Where W is the total number of repairs Constraint on spare part cost associated
of all the machines and Q is the total number with the maintenance of each machine.
of repairs of each machine. From the above
relationships the Table 2 is generated. Also 20.99X1 + 24.5X2 + 8.75X3 + 27.99X4 +
from the data collected Table 3 estimates the 43.73X5 + 83.96X6 + 69.96X7 + 10.5X8 +
value of number of repair in a year, 26.24X9 ≤ 850.69.
maximum number of hours available for
repair in a year and the percentage Constraint on depreciation cost
production hour available for each machine. associated with the maintenance of each
machine.
The objective of the linear program is to
determine the crew size that will maximize 826.97X1 + 432.4X2 + 137.83X3 + 344.57X4
the effectiveness of maintenance thereby + 344.57X5 + 826.97X6 + 344.57X7 +
maximizing the effectiveness of the 413.49X8 + 344.57X9 ≤ 9027.78.
production operation.

International Journal of The Computer, the Internet and Management Vol.16. N.o.2 (May-August, 2008) pp 26 -34

29
B. Kareem and A.A. Aderoba

Table 2: Maintenance Cost Analysis.

Machine 1 2 3 4 5 6 7 8 9 Max.
Available
cost/hr.
Manpower 13.49 15.74 5.62 17.90 28.10 53.96 44.97 6.75 16.86 347.22
cost/hr.
Spare part 20.99 24.50 8.75 27.99 43.73 83.96 69.96 10.50 26.24 850.69
cost/hr.
Depre- 826.97 432.40 137.83 344.57 334.57 826.94 344.57 413.49 344.57 9027.78
ciation
cost/hr

Table 3: Production and Maintenance Hour Analyses

Machine Number of repairs Max. hrs. available % Production hrs. Available/

S/N in a year for repair in a year production cycle.
1 12 12 99.8
2 4 14 99.8
3 2 5 99.9
4 5 16 99.8
5 5 25 99.7
6 12 48 99.4
7 5 40 99.6
8 6 6 99.9
9 5 15 99.8

Constraints on the maximum hour Subject to:

available for maintenance in each
maintenance cycle, X1 ≤1, X2 ≤2, X3 ≤2.5, 13.49X1 + 15.74X2 +5.62X3 + 17.90X4 +
X4 ≤ 3.33, X5 ≤ 5, X6 ≤4, X7 ≤8, X8 ≤1, X9 28.1X5 + 53.96X6 + 44.97X7 + 6.75X8
≤3 +0.998X9 ≤ 347.22
20.99X1 + 24.5X2 + 8.75X3 + 27.99X4 +
Non-negativity, X1≥0, X2≥0, X3≥0, … 43.73X5 + 83.96X6 + 69.96X7 + 10.5X8
X9≥0 +26.24X9 ≤ 850.69.
826.97X1 + 432.4X2 + 137.83X3 +
The general linear program is of the 344.57X4 + 344.57X5 + 826.97X6 +
form. 344.57X7 + 413.49X8 + 344.57X9 ≤
9027.78.
Max. Z* = 0.998X1 + 0.998X2 + 0.999X3 X1 ≤ 1, X2 ≤ 2, X3 ≤ 2.5, X4 ≤ 3.33, X5 ≤5,
+0.998X4 + 0.997X5 + 0.994X6 X6 ≤ 4, X7 ≤ 8, X8 ≤ 1, X9 ≤ 3
+0.996X7 + 0.999X8 + 0.998X9 X1, X2, X3, X4, …,X9 ≥0

32
 Linear Programming based Effective Maintenance and Manpower Planning Strategy: A Case Study

The linear program model generated machines. Invariably, only two crews are
was solved using software called the needed in this maintenance period. Also
Quantitative System for Business, QSB during the second maintenance cycle,
(version 3.0). The main advantage of this machines one, two and six broke-down and
software is that it does not only generate the the number of maintenance crew needed to
result, but also does the sensitivity analysis perform the maintenance job on machine,
of the optimal solutions. m/c one is one, on machine number two is
two and on machine number six is one.
4. Results and Discussion Invariably, four maintenance crews are
needed to effectively carry out the
Solving the linear programming model maintenance job during this maintenance
with QSB software, the optimal results with cycle. Following the trend of this Table, it
its objective function co-efficients sensitivity can be seen that the cycle that having the
analysis including opportunity cost are given highest number of machine breakdown (peak
in Table 5. From Table 4, which shows the maintenance period) and also requiring the
frequency of machine breakdown, Table 6 is highest number of maintenance crews is
developed to indicate how maintenance crew cycle period six (vi). At this cycle eight
must be allocated based on the results machines broke-down and twenty-crews will
obtained from the optimal solution of the be needed to carry out peak maintenance
linear programming model developed (Table jobs, instead of maximum of six crews
5). During the first maintenance cycle, only available in the section. This indicates a
machines machine number one and six shortage of 14 crews during the peak
broke-down and only one crew is needed to maintenance period.
perform the maintenance job in each of these
Table 4: Frequency of Machine Breakdown per Cycle
Machine i ii iii iv v vi vii viii ix x xi xii
1 a a a a a a a a a a a a
2 a a a a a a
3 a
4 a a a a
5 a a a a
6 a a a a a a a a a a a a
7 a a a a
8 a a a a
9 a a a a
Table 5: Sensitivity Analysis of the Optimal Results
Variable (Optimal crew) Minimum Original Maximum
X1 (1) 0.2988 0.998 Infinity
X2 (2) 0.3486 0.998 Infinity
X3 (3) 0.1245 0.999 Infinity
X4 (4) 0.3934 0.998 Infinity
X5 (5) 0.6225 0.997 Infinity
X6 (0) Infinity 0.994 1.1952
X7 (3) 0.8283 0.996 1.5952
X8 (1) 0.1494 0.999 Infinity
X9 (3) 0.3735 0.998 Infinity

International Journal of The Computer, the Internet and Management Vol.16. N.o.2 (May-August, 2008) pp 26 -34

33
 Linear Programming based Effective Maintenance and Manpower Planning Strategy: A Case Study

Table 6: Optimal Allocation of Crews to Machines

Machine Maintenance Cycle

i ii iii iv v vi vii viii ix x xi xii
1 1 1 1 1 1 1 1 1 1 1 1 1
2 - 2 - 2 - 2 - 2 - 2 - 2
3 - - - - - - - 3 - - - -
4 - - 4 - - 4 - - 4 - - 4
5 - - 5 - - 5 - - 5 - - 5
6 1 1 1 1 1 1 1 1 1 1 1 1
7 - - 3 - - 3 - - 3 - - 3
8 - - 1 - - 1 - - 1 - - 1
9 - - 3 - - 3 - - 3 - - 3

From this, it can be seen that the number there is a change in the production capacity,
of crews needed to solve the maintenance which is equivalent to the production
problem that may arise at any point in time capacity per gain (reduction) in waiting time
as derived from the optimal result without of the machine. The same condition applies
the defective machines undergoing any to the change in production capacity per
unnecessary delay is twenty. From the maintenance cycle of all the remaining
optimal results, Table 7 was gotten which machines. It can be seen also that the change
shows the old and optimal waiting times. in production capacity per gain (reduction) in
From Table 8, it can be seen that there is no waiting time of machines is negligible when
significant change in production capacity per compared with the original (old) production
maintenance cycle of Machine number one, capacity per maintenance cycle of the
since there is no change in the waiting time machines.
of the machine, but in machine number two,

Table 7: The Old and Optimal Waiting Table 8: Production Capacity (in tons) per
Times of Repair Jobs Maintenance Cycle

Machine Waiting Time (in min.) Machine Production Capacity (in tons) per
Old Optimal Maintenance Cycle
1 60 60 Old Optimal
2 120 60 1 2160 2160
3 150 50 2 2768 2771
4 200 50 3 11520 11523
5 300 60 4 3600 3604
6 240 240 5 5000 5005
7 430 160 6 1080 1080
8 60 60 7 3000 3003
9 180 60 8 4000 4000
9 3000 3003

International Journal of The Computer, the Internet and Management Vol.16. N.o.2 (May-August, 2008) pp 26 -34

33
 Linear Programming based Effective Maintenance and Manpower Planning Strategy: A Case Study

4. Conclusion References

Through the application of linear Allan, M., Lockyer, K., Oakland, J., (1988).
programming techniques to the analysis of Production and Operation
the manpower planning in maintenance, the Management. Pitman Publishing
following findings were made: the minimum Company. London. 421pp.
number of maintenance crew needed to solve Armstrong, M., (1979). Case Study in
the maintenance problem that may arise at Personnel Management. Kogan Page.
any point in time, in order to maximize the London. pp. 3.
effectiveness of the system is twenty (20); Bieghter, C.S., Phillips, D. T., Wilde, D.J.,
with the availability of this number of crew, (1979). Foundation of Optimization.
there will be an appreciable reduction in the Macmillan Press, London. 325pp.
waiting time of machines during Bronson, R., (1982). Shaums Outline of
maintenance as shown in Table 7; the change Theory and Problems of Operation
in production capacity per maintenance cycle Research. McGraw Hill Book
is not appreciable even with the use of the Company, Singapore. 423pp.
optimal number of crew (Table 8). These CBR-Conference Board Report., (1971).
findings can be used to predict the system Manpower Planning Evolving System.
operation and no doubt served as a very Vol.1. pp. 6-13. New York.
useful tool for future planning. Fapetu O.P., and Kareem, B., (2003). “An
Effective Manpower Planning
Since the twenty maintenance crews Approach for Maintenance
needed for peak maintenance period could be Departments: A Case Study of a Cocoa
idle at less maintenance periods, some of Processing Industry”, Ondo, Global J.
them could be transferred to work in of Pure and Applied Sci., 2: 59-67.
production section to promote efficiency and Heizer, J., Render, B. (1988). Production
reduce overall cost of production. To be able and Operations Management, Allyn
to achieve this objective maintenance crew and Bacon Inc., Massachusetts. 419pp.
must be multi-skilled in nature. These Howard, J.W. (1989). Production and
people should be trained and used as Operations Management. Allyn and
supplement whenever there is excess Bacon Inc. Massachusetts. 332pp.
production workload, but when there is John, F.C. (1970). Cost Optimization
excess maintenance workload they continue Engineering, McGraw Hill Book
with maintenance work. This is to avoid the Company. New York. 319pp.
cost of keeping idle maintenance crew(s). Joseph, P. (1987). Productivity Management,
For maintenance to be effective there is need Allyn and Bacon Inc. Massachusetts.
for experienced and skilled personnel. 412pp.
Therefore, further work should look at skill Kareem, B., and Aderoba, A.A., (2000).
vis-à-vis specified work to be done. “Development of a Queuing Model for
Improvement should be made in the Determining Effective Crew-Size in
documentation of maintenance data to allow Maintenance Operations”, Nig. J. of
easy analysis of the maintenance section in Eng. Manag., 1: 1-5.
the future. Kareem, B., and Aderoba, A.A., (2004).
“Application of Linear Programming
Model to Manpower Planning”,
Proceedings of the 2004 National

International Journal of The Computer, the Internet and Management Vol.16. N.o.2 (May-August, 2008) pp 26 -34

33
 B. Kareem and A.A. Aderoba

Engineering Conference of NSE, Warri,

pp. 164-171.
Kareem, B., (2005). “A Review of Strategies
for Manpower Planning in a Depressed
Economy”, J. of Sci. and Tech. Res., 4:
pp. 33-36.

___

34

View publication stats