214 Int. J. Productivity and Quality Management, Vol. 30, No.

2, 2020

Productivity improvement in furniture industry using

lean tools and process simulation

C. Shyam Murali and A. Prabukarthi*

Department of Mechanical Engineering,
PSG College of Technology,
Coimbatore, 641004, India
Email: shyammuralic@gmail.com
Email: prabukarthi.arumugam@gmail.com
*Corresponding author

Abstract: The objective of the study is to improve productivity of a furniture

manufacturing company. Demand forecasts play a crucial role in productivity,
improvement and production planning analysing the problems associated with
it and there by tackling over production and shortage. Demand forecasting was
analysed using three (simple moving average method, weighted moving
average method, seasonal regression method) different methods in which
seasonal regression was found to be more accurate for the current scenario.
This method reduces the deviation of forecast by 17.37% which will result in
better visibility thus by improvement in productivity. Preliminary survey
showed that there were some processes that lead to overall increase in
production lead time and industry is not keen in operating the second shift
particularly in painting department, which leads to under-utilisation of
resources. This research work opted for an exploratory study using the time
study analysis and value stream mapping (VSM) to identify bottleneck
processes. Brainstorming session was conducted to plot cause and effect and
there by priorities the causes to tackle. Simulation analysis was performed to
understand the utilisation of man-machine using the current and alternative
method. The proposed methods improve the operator utilisation to 54% from
29% and output by 50%.

Keywords: mean absolute deviation; MAD; mean square error; MSE; value
stream mapping; VSM; simulation.

Reference to this paper should be made as follows: Murali, C.S. and

Prabukarthi, A. (2020) ‘Productivity improvement in furniture industry using
lean tools and process simulation’, Int. J. Productivity and Quality
Management, Vol. 30, No. 2, pp.214–233.

Biographical notes: C. Shyam Murali received his Master’s (ME) in Industrial

Engineering from the Department of Mechanical Engineering, PSG College of
Technology. He also holds a BTech in Mechanical Engineering. He is currently
working as a Sales and Operations Engineer at the GMARK Middle East FZC
in Sharjah, UAE. He is also a Six Sigma Black Belt Certified candidate and his
field of interest includes lean manufacturing, Six Sigma, simulation and
modelling.

A. Prabukarthi is currently working as an Assistant Professor from the

Department of Mechanical Engineering at the PSG College of Technology,
India. He obtained his BE in Mechanical Engineering, ME in Manufacturing
System and Management and PhD in Drilling of Stacked Materials. He has
about nine years of teaching experience and two years of industrial experience.

Copyright © 2020 Inderscience Enterprises Ltd.

 Productivity improvement in furniture industry using lean tools 215

He has authored 15 journal papers and about 25 national/international

conference papers. He has been actively involved in sponsored research
projects and consultancy works. His field of interest includes supply chain
management, lean manufacturing and machining of composite materials.

1 Introduction

Productivity improvement is one of the core strategies towards manufacturing excellence

and it also is necessary to achieve good financial and operational performance. It
enhances customer’s satisfaction and reduce time and cost to develop, produce and
deliver products and service. Productivity has a positive and significant relationship with
performance for process utilisation, process output, product costs, work-in-process
inventory levels and on-time delivery. Improvement can be in the form of elimination,
correction (repair) of ineffective processing, simplifying the process, optimising the
system, reducing variation, maximising throughput, reducing cost, improving quality and
reducing setup time. Productivity enhancement is a process to achieve higher levels of
output while consuming same or lesser amounts of input resources. Also, it is believed
that if more output level is reached in a same time period and same resources, it indicates
improved productivity. It is in this respect, that the projects designed to improve
productivity must also consider resources as a major element in the process.
In furniture manufacturing industry, it should practice a well-structured production
planning, so as to meet the demand. Both over production and underutilisation are
considered to be lean wastes. Minimisation of lean wastes leads to increase in overall
productivity of the industry and a clear-cut visibility so as to plan the production as per
schedule with minimum deviation. Various tools such as value stream mapping (VSM),
time study and method study, man machine utilisation, demand forecast, etc. can be used
to analyse the current practice and propose alternatives in order to improve in overall
productivity. Analysis of past records of demand and using proper forecast techniques
which helps out to reduce the lean wastes and it will help for effective production
planning.

2 Background

Productivity improvement is one of the core strategies towards manufacturing excellence

and it also is necessary to achieve good financial and operational performance.
Productivity has a positive and significant relationship with performance for process
utilisation, process output, product costs, and work-in-process inventory levels and
on-time delivery. Improvement can be in the form of utilisation, production planning,
correction of ineffective processing, simplifying the process, optimising the system,
reducing variation, maximising throughput, reducing cost, improving quality or
responsiveness and reducing setup time
A good forecasting system helps to decrease inventory holding costs while at the
same time reducing the probability of stock-outs and customer wait times. But in the
same time there are many challenges including
216 C.S. Murali and A. Prabukarthi

 finding the best suited forecasting techniques for the scenario

 minimising the error in the forecast as much as possible
 finding/developing a relatively inexpensive easy to maintain forecasting.
This work consists of analysing the current sales or demand trend and finding the
deviation from the actual and forecasted value. Demand forecasting is an important issue
for manufacturing companies. Several decision-making processes need accurate forecasts
in order to choose proper actions relevant to production planning, sales budgeting, new
product launches, promotion planning, etc. For this reason, over the years, practitioners
and academics have devoted particular attention to how forecasting can be improved to
increase forecast accuracy (Wright et al., 1986; Armstrong, 2001; Caniato et al., 2002).
Sales forecasting plays a pivotal role in decision-making in relation to manufacturing
scheduling and inventory replenishment. Inaccuracies in forecasting can mean excess
inventories or lost sales, so it is no surprise that Mahmoud et al. (1988) and Dalrymple
(1987) report surveys that show accuracy as the most important criterion in selecting a
forecasting strategy. This may lead to the expectation that computer-based software
would be used to aid the forecasting effort. In order to tackle the overproduction or
shortage scenario that happens in the furniture industry various forecasting methods are
used in which mean absolute deviation (MAD) and mean square error (MSE) is used as a
measuring tool to ensure the accuracy of the forecasting.
Time study is one of the oldest fundamental method ever used which results in
increased productivity. Since the research is focus on the time study from all aspects, so
the history of the time study must be analysis and how it develops into the most useful
tools in manufacturing industry. Work chart is a tool that can help to identify problems
and provide solutions through describe the work activities in a systematic and clear step
such as transportation, operation, inspection, inventory and delay. The application of
science to business problems, with the use of time-study methods in standard setting and
the planning of work, was pioneered by Frederick Winslow Taylor, with emphasis on fair
day work. The study reveals that tremendous amount of coordination efforts was put with
managerial positions to prepare the path for human relations to supersede scientific
management in terms of literary success and managerial application. The very fact that
the oral management was converted to scientific management, with all facts, data analysis
to enhance productivity irrespective of the physiological cost to the worker. Cadbury and
Marshall heavily criticised Taylor and pervaded his work with subjectivity. Cadbury in
reply to Thompson stated that under scientific management employee skills and
initiatives are passed from the individual to manage, they relied heavily on individual
interpretations of what workers actually do and Gilbreths were charged with the task of
proving that motion study particularly, and scientific management generally, increased
industrial output in ways which improved and did not detract from workers mental and
physical strength.
VSM is one of the key lean tools used to identify the opportunities for various lean
techniques. VSM is a tool which enables a company to map the process flow that helps in
identifying various factors like:
 value added (VA) time (time taken for producing the end product)
 non-value added (NVA) time (time taken which do not contribute to the production
of end product)
 Productivity improvement in furniture industry using lean tools 217

 cycle time (time required to perform a process)

 takt time (maximum amount of time in which a product needs to be produced in
order to satisfy customer’s demand).
Chen (2012) proposes a VSM-based lean production system for Chinese enterprises to
help them deploy lean production systematically, which can make them have an overall
look at total efficiency, tell them where they are, where they want to go and map a route
to get there. This paper will help them eliminate roots of wastes, rearrange overall value
stream better and increase the competitive ability of Chinese enterprises.
In Goriwondo and Mhlanga (2012), the paper details the use of the VSM tool in
reducing waste in bread manufacturing for a company in Zimbabwe. The case study
shows how the VSM tool was used to identify and reduce defects by 20%, unnecessary
inventory by 18% and motion by 37%. It incorporates waste relationship ranking and the
importance of management commitment in waste reduction. The study set to use the
VSM tool in troubleshooting waste generated in bread manufacturing and identify ways
of reducing this waste while at the same time increasing the proportion of the processes
that add value to the product. This was achieved through the development of the future
state map which has an increased throughput of 16%.
Ishikawa (1985) has described cause and effect diagram which rightly helps to
identify causes related to the effect, using the brainstorming technique.
Khalafi and Raissi (2014) compared the bottleneck’s present in different layouts
which are modelled by using enterprise dynamic simulation software. Critical parts were
identified in aerators and valves manufacturing layout and the best among the layouts
was suggested to reduce lead time. Leporis and Kralova (2010) examine the bottleneck
detection methods. The statistical data is obtained from witness simulation software and
compared with the detection methods to adopt as per the requirements.
Simulation is a computer-based modelling of a real production system. Simulation
allows organisations in the manufacturing industry to analyse and experiment with their
processes in a virtual setting, reducing the time and resources associated with processes.
Simulation offers a powerful tool to support the understanding the current scenario with
respect to the throughput and with respect to the man machine utilisation. Ong (2007)
discussed the deficiency of traditional static methods and how simulation can be applied
to overcome these deficiencies in order to establish an optimal man-machine ratio. It also
discussed using simulation to facilitate decision-making on headcount movement from
non-constraint tool to constraint tool to boost productivity of constraint tool, and
ultimately increase overall factory output.

3 Purpose and theoretical framework of the paper

In furniture industry, production planning plays a vital role and the current forecasting
practice leads to inventory in some period or reverse stock in some period. Demand
estimation is very crucial for a manufacturing company. Improper forecasting results in
lack of visibility in production planning and demand estimation. During March, the
forecast deviation reaches up to 25% from actual, which makes all production planning
imperfect. Therefore, analysis and improvement for the forecasting will be done in this
project in order to enhance the visibility and production planning in the industry.
218 C.S. Murali and A. Prabukarthi

Production lead time is high (2,552 min) due to bottleneck processes. Bottleneck
identification is a major task so as to tackle the low throughput in the bottleneck process.
As sales trend was going up in order to make industry fit for fulfilling the customer’s
demand it is required to identify and control the bottleneck.
Industry is not keen in operating the second shift particularly in painting department,
which leads to poor utilisation of resources. Operator utilisation observed as 46.73% per
cycle and machine utilisation 83.89% per cycle in painting department where considering
a complete shift operator utilisation comes down to 28.2% and 29.1%. In second shift
machine is not utilising because of operator unavailability. So as to improve utilisation
and improve productivity, by improving the average daily output with the same
workforce. Gap analysis is carried out shown in Table 1. The purpose of the gap analysis
is to provide project teams with a format in which to do the following:
 compare the best practices with the processes currently in the industry
 determine the ‘gaps’ between current practices and the identified best alternatives
 select the best alternative practices to implement.
Table 1 Gap analysis

Action
Current situation Future state Expected outcome
plans/proposals
1 Forecasting Better alternative Comparative analysis Maximum
deviation up to 25% with minimal with three deviation less than
from the actual deviation alternatives, MAD 10%
and MSE comparison.
2 High production Throughput Time study, VSM and Bottleneck
lead time due to improvement in cause and effect identification
bottleneck analysis bottleneck processes diagram
3 Underutilised Resource balancing Simulation Double the current
resources in operator
painting department utilisation, increase
output at least by
10%

The sales data for various products produced by the industry has been analysed for a
12-month period. The data has been summarised into a Pareto chart as shown in
Figure 1. The Pareto chart shows that among the number of products 58.5% of the sales
volume are from finished wooden boards and customised furniture and both are made out
of finished boards. So, this project focus on production and forecasting practices for
boards. All the analysis and improvements done based on the wooden boards they
produce.
The sales data for various products produced by the industry has been analysed for a
12-month period. The data has been summarised into a Pareto chart as shown in Figure 1.
The Pareto chart shows that among the number of products 58.5% of the sales volume are
from finished wooden boards and customised furniture and both are made out of finished
boards. So, this project focuses on production and forecasting practices for boards. All
the analysis and improvements are done based on the wooden boards they produce.
 Productivity improvement in furniture industry using lean tools 219

Figure 1 Sales volume for 12 months (see online version for colours)

4 Data collection and analysis

4.1 Forecasting techniques used for demand estimation

There are number of forecasting techniques that are widely used according to the data to
be forecasted. A rubber wood furniture manufacturing company where the availability of
raw materials and the demand from the customer end both will follow a seasonal trend.
Where the demand trend may not be a proper seasonal trend but it will follow an upward
trend with a seasonal variation. Here the discussions include comparison between three
forecasting methods and finding which one is best suited for the scenario.

4.1.1 Comparative analysis

The below graphs shows the comparison of the simple three-month moving period
average forecast method with the actual sales which is being practiced by the industry
currently. From Figure 2, it can be observed that the forecasted values are always lags the
actual trend which results an overproduction and under production in some period. The
deviation observed is quite large which clearly indicates that it is not a suitable
forecasting method for a sales trend as such.
Compared to SMA, WMA is giving close value in deviation with actual. But at the
same time the trend is lagging in WMA also with lesser deviation compared to SMA.
Below graphs show WMA with different weightages with set of (0.5, 0.3, 0.2) and (0.7,
0.2, 0.1). Compared to SMA, the deviation from the actual sales is lesser for WMA, as
shown in Figure 2. As the sales history follows a combination of increasing trend with a
kind of seasonal behaviour, the seasonal regression forecasting method is carried out and
found the result. The resulting forecast trend is somehow follows the similar trend with
the actual sales trend with much lesser deviation while compared to the other forecasting
techniques. Figure 2 graph shows the comparison of all the three forecasting techniques
used here with actual trend which clearly indicates that the seasonal regression is much
220 C.S. Murali and A. Prabukarthi

closer to the actual as well as no lagging in trend. Each group of six months categorised
as H1 to H4 which is showing the behaviour of the sales trend in different time period.

Figure 2 Forecasted vs. actual sales (see online version for colours)

H1 H2 H3 H4

Figure 3 MAD comparison in H1 (see online version for colours)

Figure 4 MSE comparison in H1 (see online version for colours)

 Productivity improvement in furniture industry using lean tools 221

Figure 5 MAD comparison in H2 (see online version for colours)

Figure 6 MSE comparison in H2 (see online version for colours)

Figure 7 MAD comparison in H3 (see online version for colours)

Figure 8 MSE comparison in H3 (see online version for colours)

Figure 9 MAD comparison in H4 (see online version for colours)

222 C.S. Murali and A. Prabukarthi

Analysing the graphs in Figure 3 to Figure 10, are the pictorial representation of the
MAD and MSE from the actual sales in H1 to H4. From the graphs, it clear that among
each group the deviation is lesser for Seasonal regression. Even if the trend is normal or
peak seasonal regression is best suited for the current scenario with the current data
collected.

Figure 10 MSE comparison in H4 (see online version for colours)

Among the forecasting techniques mentioned, minimum deviation or error are plotted by
seasonalised time series forecast, i.e., close to the actual sales (Gaither, 2011). By
calculation:
a  18705804.57
b  209625.0901
Yd (forecasted sales)  a  bX

Seasonalised forecast  Seasonality index  Yd

Demand forecasting is analysed using three (simple moving average method, weighted
moving average method, seasonal regression method) different methods in which
seasonal regression is found to be more accurate for the current scenario. Based on the
results from seasonal regression model, improvement of reducing the deviation of error
forecast by 17.37% which will result in reduction of over production and shortage. A set
of six months wise comparison also made in order to infer the suitable forecasting
methods for peak and normal demand trend. On study of the various methods applied, the
seasonal regression forecasting method has found to be the most suitable by comparison
of the MAD and MSE which studies the deviation between the forecasted and actual
values of demand.

4.2 Time study

Time study is work measurement technique for recording the times of performing a
certain specific job or its elements carried out under specified conditions, and for
analysing the data so as to obtain the time necessary for an operator to carry it out at a
defined rate of performance. As in method study, the first step in time study is to select
the job to be studied. Time study is carried out and consolidated and used for the input
parameters for VSM. The consolidated time including the segregated form as like VA
time and NVA time is as shown in Table 2.
 Productivity improvement in furniture industry using lean tools 223

Figure 11 Current state value stream map (see online version for colours)
224 C.S. Murali and A. Prabukarthi

Table 2 Time consolidation

VA time Total C/T VA NVA

Process Operators Shifts
(minutes) (sec) (sec) (sec)
Top and bottom planer 2.04 467 122 344 2 2
Rip saw operation 1.91 416 115 301 4 2
Finger joining 1.63 371 97.8 274 3 2
Moulding 11.4 1,836 684 1,152 3 2
Composing 3.75 742 225 512 2 2
Sanding operation 1.27 322 76.2 185 2 2
Panel saw operation 2.17 425 130 295 2 2
Sealing and lackering 35.2 2,677 2,112 528 2 2

4.3 Value stream mapping

A value stream map is an end-to-end collection of processes/activities that creates value
for the customer. A value stream is all the actions (both VA and NVA) currently required
to bring a product through the main flows essential to every product:
 the production flow from raw material into the hands of the customer
 the design flow from concept to launch.
It always starts at customer demand and works back through documenting all processes
and collecting data which is required for manufacturing of a product. It shows the map
‘value stream’ containing all the data such as work in process, cycle time, lead time,
number of equipment’s, operators and information flows. In this particular case, even
though we find out that the process can meet the existing demand but a slight increase in
it would lead to occurrences of bottlenecks and thereby will be a hindrance to the entire
production process. Thus, productivity is enhanced by knowing the bottlenecks and by
eliminating them.
All the process incorporates with production of wooden boards are analysing and
used to plot VSM. Time study consolidation with considering operators and decoupling
inventory duration or quantities between processes. Figure 11 shows the current state
VSM from which bottleneck process have to be identified.

4.3.1 Takt time calculation

Hours per shift  8 hours

Planned loss  75 minutes

Net available time(minutes/day)  8 hours shift  planned loss

 480  75  405 minutes per shift

Net available time(minutes/week)  2  405  6  4896 minutes

Customer demand for 1 week  250 units

 Productivity improvement in furniture industry using lean tools 225

Net available time

Takt time   4896 / 250  19.5 minute/unit
Customer demand

From the work balance graph Figure 12 showing cycle time vs. takt time it implies the
painting process includes sealing and lackering is the bottleneck. All the other processes
including moulding and composing which is having comparatively high cycle time also
falling under the takt time. That is nothing but in each 19.5 minute, customer needs a unit
and it is blocked by the painting process.

Figure 12 Cycle time vs. takt time (see online version for colours)

4.4 Cause and effect diagram

Brainstorming session is conducted with detailed discussion on low throughput in
painting department. The participants are production manager, shop floor engineers and
experienced operators. Brainstorming session reveals various causes and sub causes the
results in low throughput. Figure 13 shows the cause and effect diagram for low
throughput.

Figure 13 Cause and effect diagram for low throughput in painting department
226 C.S. Murali and A. Prabukarthi

Mainly, the problem faced is lack of skilled operator to run the night shift. But the same
time in shift 1 the utilisation of labour is less because of few essential NVA time like
drying time. So, work assignment and resource balancing are the suggestions made after
brainstorming sessions at the painting department. It also decided to concentrate on work
balancing and utilisation of operator.

4.5 Simulation
Simulation experiments were performed for the bottleneck process to calculate the
man-machine utilisation and output. It is already found that process is consuming high
essential NVA time which cannot be eliminated. An alternative is being proposed with
reassigning the operator who will operate two machines instead of one machine with
considering allowances according to ILO (2005) standards.
Once all the required data and the shop floor are identified, the details are entered into
commercial simulation software. A number of assumptions are made during the
development of the simulation model:
 the process flow for the simulation is continuous

 production is in an ideal situation, in which no breakdown occurs and no product

quality problem exists

 an equal ratio of volume exists for each part

 all the parts arrive randomly and the parts will go into a certain machine when
available

 the operator moving time between workstations is assumed to be constant

 the machine cycle times are assumed to be the same for both machine with same type
of part

 the production plant runs with time table as per their work schedule.
Three measurement criteria were considered: output, machine utilisation, and operator
utilisation. Output is measured both per shift and per day, and compared both simulated
and actual for current as well as proposed scenario. Machine utilisation measures the
machine usage intensity and is computed as the percentage of busy hours that a machine
runs to complete the given tasks. The data can be acquired from the statistical report
generated by the simulation software. Operator utilisation reflects the overall
performance or productivity of the process and refers to the percentage of busy hours that
an operator requires to complete given tasks. The data can be obtained from the statistical
report generated by the simulation software.

4.5.1 Man machine chart

A man-machine activity chart is a chart used to describe or plan the interactions between
workers and machines over time. The colour green indicates utilisation and the colour red
indicates idle. As the name indicate, the chart deals with the criteria of work elements and
their time for both the worker and the machine. This chart is useful to describe any
repetitive worker-machine system. The chart is prepared for the bottleneck process, i.e.,
 Productivity improvement in furniture industry using lean tools 227

painting department. There are essential NVA processes plays a major part in the area of
productivity improvement painting department the process is man-machine chart for a
cycle is found out and analysed. Table 3 shows the man-machine chart for a cycle in
painting department, process including sealing and lacquering.
Table 3 Man machine chart (see online version for colours)

Activities Time (minutes) Man Machine

Task preparation Order receipt 1.43
and loadig Part transportation 1.94
Mixing colour 6.00
Loading of paint 3.60
Loading part to work table 3.08
Painting Sealing 1st coat 7.69
operation Drying time 25
Sanding operation 3.50
Painting Sealing 2nd coat 5.50
operation Drying time 20
Sanding operation 2.50
Painting Lackering 5.70
operation Drying time 20
Unloading and Unloading 3.18
moving Moving to stock keeping area 2.60

The colour red indicates idle time and green is utilisation. The utilisation of man and
machine for a cycle is calculated and this gives the indication that utilisation of man will
be less for a shift of process also. Simulation for a complete shift is being done using
commercial software.

Figure 14 Man-machine utilisation per cycle (painting department), (a) machine utilisation
(b) operator utilisation (see online version for colours)

(a) (b)

Man machine chart for a cycle is found out and analysed. Figure 14 shows the pictorial
representation of the utilisation of both man and machine. Operator (man) utilisation is
less than 50% for a cycle and it is even lesser for a complete shift. A shift of process is
being simulated using commercial simulation software and that will discuss further.
228 C.S. Murali and A. Prabukarthi

4.5.2 Simulation using commercial software

Simulation for a complete shift with current timings and practices shows in Figure 15
much different result in utilisation of operator (man) compared to utilisation in one cycle
and it is found that operator utilisation is 29.1% in a one shift of period. It is a two
machine two operator case with simulated average output is 30 units. In the actual
scenario output varies from 28 to 34. In that 34 is an extreme case obtained by working a
longer shift 1.

Figure 15 Two machine two operators shift 1 (current practice) (see online version for colours)

In this case machine is being utilised by 89.5% and 91.1% where the operator utilisation
is 28.2% and 29.1% respectively. This directly projects that up to 70% time operator is
idle. As already discussed a high amount of essential NVA time can be utilised properly
with a same operator. An alternative to the system is proposed with one operator who will
handle two machines alternatively. This proposal leads to an improvement in utilisation
of operator.
The simulated result shown in Figure 16 shows that machine is being utilised by
88.2% and 84.4% where the operator utilisation is 54% and average output is about
24 units only but the main advantage with this proposal is that a night shift can be
introduced with the extra operator and that tackles the causes like large essential NVA
time, high idle time of operator, insufficiency of operator in running night shifts, etc. that
are founded out for low throughput by brainstorming session.
The proposed alternative is to run the process for two shifts with the simulated
average output being 54. With better utilisation of man and machine which is presented
in Figure 17. From the pilot run result of this proposal it is found that output per day (two
shifts) is 45 units, 21 units from shift 1 and 24 units from shift 2.
 Productivity improvement in furniture industry using lean tools 229

Figure 16 Two machine one operator shift 1 (alternative) (see online version for colours)

Figure 17 Two machine one operator (two shift) (see online version for colours)

5 Results and discussion

This intense study on current operations and forecasting practices is carried out. Mainly,
three forecasting namely simple moving average, weighted moving average and seasonal
regression methods. Initially the maximum deviation from the actual with SMA method
practice is getting 25%. The comparative analysis shows that the seasonal regression
230 C.S. Murali and A. Prabukarthi

holds a maximum deviation of 7.6%, which is the minimum compared to other methods.
Although the sales trend is not a pure seasonal demand, it follows an upward trend with
some peaks in certain period, mainly in the month of March. With implementation of best
alternative to the simple moving average method, i.e., seasonal regression, the results
indicate that it is reached to 3.5% in March 2017. An error bar with ± 5% is assumed for
the forecasted value which is shown in Figure 18, it shows that the actual value is coming
inside the ± 5%. This makes obvious improvement in visibility and better production
planning.

Figure 18 Actual vs. forecasted comparison with error bar of 5% (March 2017) (see online
version for colours)

Standard time calculation done with time study and this is being used to plot current state
VSM for the production of wooden boards. Using VSM of current state with bottleneck is
identified, that is painting department. Brainstorming session with production manager,
floor engineer, supervisors and operators and certain quality outcome including plotting
the causes and effects diagrams for the low throughput in the bottleneck process. Mainly
focused on the underutilisation of operator, high essential NVA process time and
insufficient operators for night shift.
Simulation using commercial software used to analyse current utilisation and output
for one complete shift with two operators working on two different machines.

Figure 19 Average output per day in current vs. alternative practice (painting) (see online version
for colours)
 Productivity improvement in furniture industry using lean tools 231

The simulated output of alternative with single operator for two machines is found out
better in utilisation of operator hour from 29.1% to 54% where output per shift varies
from 30 to 24 units. But this proposal holds an added advantage with an extra labour who
is assigned to work in night shift which gives a simulated result of average output per day
54 units, shows in Figure 19. From the pilot run with this proposal gives an average
output of 45 units a day which is a 50%–60.7% increase in output. Finally, the overall
idle time of an operator get reduced from 71.8% to 46% as shown in Figure 20.

Figure 20 Idle time in current vs. alternative practice (painting) (see online version for colours)

6 Conclusions

Purpose of the study is to improve productivity of the furniture industry. The following
findings are the major outcome from the study:
 Lean wastes were identified and tackled.
 Underutilisation of operators were improved by proposing that the same employee to
operator two machines which can help in reducing, idle time or essential NVA by
utilising that particular period for VA processes.
 The demand for the furniture’s were studied by conducting an analysis using three
forecasting techniques namely, simple moving average, weighted moving average
and seasonal time series. Seasonal time series forecast method provides minimum
deviation from actual when compared with two forecasting methods. Forecasting
analysis helps in effective production planning which result in minimising the
investment on inventory.
 Processes in the paint department have been identified as bottleneck process and
simulation-based study has been carried out to identify the alternative methods in
order to improve the throughput. Based on the simulation study the expected output
was 54 units. From the pilot run it is realised that the 45 units can be produced.
 The percentage of output has been improved from 50% to 60.7% when compared to
earlier production rate.
232 C.S. Murali and A. Prabukarthi

References
Armstrong, J.S. (2001) Principles of Forecasting: A Handbook for Researchers and Practitioners,
Kluwer, Boston.
Arun, K.S. and Rajenthirakumar, D. (2016) ‘Reducing work in progress inventory in a valve
industry using value stream mapping’, International Journal of Engineering Research, Vol. 5,
No. 4, pp.308–310.
Barnes, R.M. (1980) Motion and Time Study Design and Measurement of Work, Quinn-Woodbine,
Inc. Wiley, New York.
Belokar, R.M. and Kharb, S.S. (2012) ‘An application of value stream mapping in automobile
industry: a case study’, International Journal of Innovative Technology and Exploring
Engineering, Vol. 1, No. 2.
Cadbury, E. (1914) ‘Some principles of industrial organization: the case for and against scientific
management’, Sociological Review, Vol. 7, pp.99–125.
Caniato, F., Kalchschmidt, M., Ronchi, S., Verganti, R. and Zotteri, G. (2002) ‘Forecasting demand
fluctuations due to promotional activities: a case in the fresh food industry’, in Proceedings of
the POMS Conference, San Francisco.
Charolaet, M.B. (2015) ‘Productivity improvement by use of time study, motion study, lean tool’s
and different strategy for assembly of automobile vehicles’, IJSRD – International Journal for
Scientific Research & Development, Vol. 3, No. 2.
Chen, J.C. and Cox, R.A. (2012) ‘Value stream management for lean office – a case study’,
American Journal of Industrial and Business Management, Vol. 2, No. 1, pp.7–29.
Chen, L. (2012) ‘The application of value stream mapping based lean production system’,
International Journal of Business and Management, Vol. 12, No. 6.
Chowdhury, S. and Haque, K.A. (2015) ‘Implementation of lean strategies in a furniture
manufacturing factory’, IOSR Journal of Mechanical and Civil Engineering, Vol. 12, No. 1,
pp.45–50.
Dalrymple, D.J. (1987) ‘Sales forecasting practices: results of a United States survey’,
International Journal of Forecasting, Vol. 3, Nos. 3–4, pp.379–391.
Durana, C. and Cetindereb, A. (2016) ‘Productivity improvement by work and time study
technique for earth energy-glass manufacturing company’, Procedia Economics and Finance,
Vol. 26, pp.109–113.
Gaither, N. (2011) Operations Management, Cengage Learning, Southwestern College Publication.
Gill, P.S. (2012) ‘Application of value stream mapping to eliminate waste in an emergency room’,
Global Journal of Medical Research, Vol. 12, No. 6.
Godoy, L.P. (1998) ‘Quality and productivity improvement in a custom-made furniture industry
using kaizen’, Computers Industrial Engineering, Vol. 31, Nos. 1–2, pp.471–474, Elsevier.
Goriwondo, W.M. and Mhlanga, S. (2012) ‘Use of the value stream mapping tool for waste
reduction in manufacturing: case study for bread manufacturing in Zimbabwe’, International
Conference on Industrial Engineering and Operations Management.
Haque, K.A. (2015) ‘Implementation of lean strategies in a furniture manufacturing factory’, IOSR
Journal of Mechanical and Civil Engineering, Vol. 12, No. 3, pp.45–50.
ILO (2005) Introduction to Work Study (Indian Adaption), pp.271–297.
Ishikawa, K. (1985) What is Total Quality Control, Lu, D.J. (Trans.), Prentice Hall, New Jersey,
ISBN: 0139524339.
Jilcha, K. and Berhan, E. (2015) ‘Workers and machine performance modeling in manufacturing
system using arena simulation’, Journal of Computer Science & System Biology, Vol. 8,
pp.185–190.
Khalafi, S. and Raissi, S. (2014) ‘Industrial valves production line bottleneck analysis: a computer
based simulation approach’, International Journal of Scientific & Technology Research,
Vol. 3, No. 6.
 Productivity improvement in furniture industry using lean tools 233

Leporis, M. and Kralova, Z. (2010) ‘A simulation approach to production line bottleneck analysis’,
International Conference Cybernetics and Informatics, 10–13 February.
Mahmoud, E., Rice, G. and Malhotra, N. (1988) ‘Emerging issues in sales forecasting and decision
support systems’, Journal of the Academy of Marketing Science, Vol. 16, Nos. 3–4, pp.47–61.
Nallusamy, S. and Saravanan, V. (2016) ‘Lean tools execution in a small scale manufacturing
industry for productivity improvement – a case study’, Indian Journal of Science and
Technology, Vol. 9, No. 35, pp.1–7.
Ong, H.H. (2007) ‘Establishing man-machine ratio using simulation’, Proceedings of the Winter
Simulation Conference.
Parthiban, P. (2015) ‘Productivity improvement in shoe making industry by using method study’,
IOSR Journal of Mechanical and Civil Engineering, pp.1–8, ISSN: 2278-1684.
Radharamanan, R. and Godoy, L.P. (2010) ‘Quality and productivity improvement in a custom-
made furniture industry using kaizen’, International Conference on Computers and Industrial
Engineering, Vol. 31, Nos. 1–2, pp.471–474.
Turkyilmaz, A. (2013) ‘Value stream mapping: case study in a water heater manufacturer’,
International Journal of Supply Chain Management, Vol. 2, pp.2–5, ISSN: 2050-7399.
Verma, N. and Trivedi, P. (2015) ‘Productivity improvement in assembly line of an automobile
industry’, IOSR Journal of Mechanical and Civil Engineering, Vol. 12, No. 4, pp.1–6.
Wright, D.J., Capon, G., Page, R., Quiroga, J., Taseen, A.A. and Tomasini, F. (1986) ‘Evaluation
of forecasting methods for decision support’, International Journal of Forecasting, Vol. 2,
No. 2, pp.139–152.
Yusoffa, N. (2012) ‘Work measurement for process improvement in the car seat polyurethane
inject ion manufacturing line’, Procedia Engineering International Symposium on Robotics
and Intelligent Sensors, Elsevier.