TIME AND METHOD
Predetermined standard times
The Predetermined Standard Time systems are based on
the basic principle that every elementary movement /
activity requires practically the same time, with the same
working conditions and if performed by a sufficiently
skilled executor
Times are expressed in the particular unit TMU (Time
Measurement Unit)
• 1 TMU = 0.00001 hours = 0.0006 min = 0.036 sec
• 1 hour = 100,000 TMU
For the calculation of the Standard Times, a correction
coefficient F is usually added
Predetermined standard times
Steps of the method to Predetermined Standard Times:
1. Breakdown of the work to be performed in its basic
microelements
2. Identification in the appropriate tables of the TMU values
related to the micro-movements
3. Adjustment of values through corrective factors
4. Execution of the sum of the values of all the micro-
elements to be performed to carry out the work
5. Determination of the total standard time
Predetermined standard times
There are several families and subfamilies of methods /
systems for the calculation of predetermined standard
times
The most common (from which derives most of the
others) is the family known as MTM (Method Time
Measurement)
The different MTM systems allow the applicability of the
method according to the diversity of user needs. The
main families are:
o Motion-based systems MTM 1
o Element-based systems MTM II (es. MTM UAS, MTM MEK,
MTM-HC)
o Activity-based systems -MOST
Predetermined standard times
The original MTM method defines the times of the main
micromovements of upper limbs, eyes and lower limbs
The 9 micromovements of the upper part of the body
o (Reach)
o (Move)
o (Turn)
o (Apply Pressure)
o (Grasp)
o (Release)
o (Position)
o (Disengage)
o (Crank)
Each movement corresponds to a table that provides the TMU according
to the boundary factors (distances to be covered, weights, shapes of
objects ..)
Tempi standard predeterminati
Tempi standard predeterminati
Motion-based (MTM 1)
MTM 1 is a very detailed and reliable system that focuses
on the analysis of the movements of the two hands
It is suitable for the study of processes:
o to another degree of repetitiveness
o very short cycles,
o when errors of a few TMUs could cause great inconveniences
in production and economic convenience
For example for brake assembly lines
Motion-based (MTM 1)
Element-based (MTM 2)
he Element-based family is a derivative of MTM-1,
corresponding to a simplification of the detected
movements and a specialization in different sectors
There are a number of subfamilies of specialization in the
sector, e.g. MTM-HC (for the healthcare industry), MTM-
C (for office work), MTMM (for microscopic work ...)
MTM UAS is a system derived from MTM-1 through
statistical processing of the tabulated data, which does
not distinguish the detail movement of the two hands
It is the result of an aggregation of the basic movements
of MTM 1 in main handling elements ,.
Suitable for processes characterized by significant
variations in the production cycle
Element-based (MTM 2)
Activity-based (MOST)
MOST (Maynard Operation Sequence Tecnique) is a
faster MTM system than previous families, because it
identifies the main activities and not the single
movements
Naturally it loses in level of detail and therefore precision
in the elaboration of the standard times MOST defines
not a series of movements, but a sequence of events /
activities that involve the movements
The basic MOST events are:
o The sequence of movement of an object
o The control sequence of an object
o The sequence of use of tools and an object
o The sequence for the use of manual cranes
Activity-based (MOST)
Alongside each sub-activity is the execution time, which
derives (as in the other methods) from standardized
tables according to different parameters (eg number of
steps within the sub-activity)
o The time indicated in the index is 1/10 of a standard TMU
o The standard time is obtained as TMU + allowance
factor, where allowance factor = increase of the standard
time for personal rest (P), fatigue (F), different delays (D)
o Usually the allowance factor is at least 15% of the
standard time calculated with MOST
General move
ABGABPA
A action distance
B body motion
G gain control
P placement
General move
represents the activity "Walk for three steps and take a
bolt from the floor, lift it up and put it in a box",
A6 B6 G1 A1 B0 P3 A0
TMU = (6 + 6 + 1 + 1 + 0 + 3 + 0) * 10 = 170 TMU = 0,102
min Tempo standard = 0,102 min * 1,15 = 0,1173 min
con allowance factor pari al 15%
Controlled move
ABGMXIA
A action distance
B body motion
G gain control
M Move controlled
X Process time
I alignment
Controlled move
indicates the activity of setting a control parameter on a
machine (eg milling machine)
A1 B0 G1 M1 X10 I0 A0
Tool use
ABGABP… ABPA
A action distance
B body motion
G gain control
P placement
… F fasten, L loosen, C cut, S surface treat, M
measure, R record, T think
Manual crane model
ATKFVLVPTA
A distanza percorsa
T trasport unloaded
K Hook-Unhook
F Free
V Vertical move
L Loaded move
P Placement
Activity-based (MOST)
Advantages
The standard times can be accurately evaluated
(differently depending on the MTM family) before
production starts
You can compare without making more alternatives on
the work cycles
The possibilities of error in the recording of times and
performances are theoretically reduced
It is easier to apply and cheaper than Time Study
systems. They are usually more easily accepted by trade
union
Disadvantages
It is practically inapplicable if the activities are not very
repetitive
In the application of more detailed families (eg MTM 1)
the division of labor into micro-operations can be very
difficult
The parameters chosen for the timing determination may
not be suitable for any work situation
Factors that could introduce variability in execution times
are potentially unlimited, therefore not all are included in
the tables (eg MTM 1 does not consider the shape of the
pieces to be moved)
Example
ABGABPA
ABGMXIA
ABGABP… ABPA
Learning curve
Definizione
The learning curve (or progress curve or learning curve) is a tool
used to design (or reorganize) production systems in
consideration of variations that occur over time as a result of the
learning phenomenon.
‘The productive efficiency of each
activity increases continuously
by repeating this activity’
A concept that, translated into appropriate mathematical models,
makes it possible to forecast with reasonable accuracy the
variation in time of learning-related quantities (and progress)
such as the unit cost of the product, the time needed to build it,
the maintenance hours necessary, etc.
Learning
Learning is the sum of:
Discrete factors: they cause a practically instantaneous and easily perceived variation of the
observed quantity
Inventions
Discoveries
applications., widespread and in short time, of innovative technologies
Continuous improvements: non-perceptible events, if the observation is superficial, due to the
areas
design
technological / technical
organization / management
Continuous improvements
Project area
o documentation on the state of the art
o product design
o better definition of operating methods
Technological / technical area
o automation
o Application of alternative technologies
o Procedures optimization
More appropriate choices of tools
o Management organizational area
o Organization of departments
o level of training
Production control
o Employment of labor
o Use of materials
o Use of energy
Continuous improvements
In order to continuously improve it is still necessary to create ideal
conditions in the company. In fact, learning depends on:
Attitude / ability to learn
o Physical adaptability
o Cultural degree
o Grounds
Characteristics of the work to be done
o complexity
o Length of cycle times
Boundary conditions
o External motivations
o Changes in situations
o Conditions related to work
Wright Model (1936)
𝑦 = 𝑎 ∗ 𝑥 −𝑏
y = measure of productivity (eg: unit cycle time, unit cost, unit weight)
a = parameter linked to the measure of productivity or productivity at the
initial time (e.g.: first piece)
x = cumulated volume of production
b = learning rate or margin of productivity margin
Wright model
Wright model
The variation in productivity is normally expressed in
percentage terms which obviously corresponds to a
precise numerical value of the parameter b
Productivity variation % Learning curve(b)
55 0,8292
60 0,7372
70 0,514
80 0,322
90 0,152
95 0,074
Wright model
How to define the parameters a and b
• Cochran
• Williams
• Baloff
• Westinghouse
Wright model
Determination of the values of parameters a and b
Cochran method:
a) Determine the value of steady-state productivity (eg:
standard productivity after n productions)
b) Estimate the percentages of improvement of each of the
activities in which the production is divided
c) Assign a weight to each activity to arrive at a weighting
average improvement rate
d) Starting from the regime productivity and thus having
estimated the rate of improvement, determine the initial
productivity a
Wright model
Determination of the values of parameters a and b
Williams method:
a) Examination of curves for similar productions and
confirmed in practice
b) The curve considered most suitable among the
examined is taken into consideration, assuming its rate
of improvement b
c) The productivity of the second production is measured
by calculating the initial productivity a
Wright model
Baloff method:
A correlation between the rate of improvement b and
productivity y is considered
Westinghouse method:
a) Like Williams, it takes a characteristic curve to determine
the rate of improvement b and productivity y
b) Like Cochran, he estimates steady-state productivity and
the amount of production to achieve it to determine initial
productivity a
Determination of the values of parameters a and b
By increasing the cumulative production, as a result of
learning rates related to the various operations, it could
generate imbalances between the stations of the line. To
avoid this problem and rebalance the lines, one can take
a cue from what Dar-El and Rubinovitz suggested
Determination of the values of parameters a and b
Dar-El and Rubinovitz method
o Using fairly simple criteria, the operations are divided into two
categories
o Phases characterized mainly by intellectual learning. These phases are
further distinguished between phases of high and low intellectual learning,
respectively with improvement rates b between 70% and 75% and rates
between 75% and 80%
o Stages characterized mainly by acquisition of manual skills. These
phases are further distinguished between phases of high and low
acquisition of manual skill, respectively with b improvement rates between
80% and 85% and rates between 85% and 90%
o The weighted curve to be used is determined
o By means of an algorithm that takes into account the different
regressions, the redefinition of the production lines is achieved as a
function of the increase in the cumulative production
Determination of the values of parameters a and b
To take into account the fact that production can be
suspended for a certain period of time and then resumed,
typical of batch production, it is necessary to consider the
possibility of forgetting, the forgetting factor. In general
terms the forgetting factor depends mainly on the time
interval between the production of a batch and the next
and the complexity of the operations.
Towill exemplified what can happen when the forgetting
factor changes.
Determinazione dei valori dei parametri a e b
Towill analysis
a) A certain quantity of identical parts is considered to be
produced in n lots of equal sizes interspersed with equal
times
b) The size of the lots is varied
c) The value of the forgetting factor expressed in
percentage terms is changed
d) The result is analyzed according to the total time
necessary for the production of the whole quantity
assumed.
Towill
PARTI / GIORNO
FO FA = 0%
t n+1 < t n
t1 t2 t3 t4
t totale
Towill
PARTI / GIORNO
FO FA = 50%
t n+1 < t n
t1 t2 t3 t4
t totale
Towill
PARTI / GIORNO
FO FA = 100%
t n+1 = t n
t1 t2 t3 t4 t totale
Other model
Curves to S
The canonical S curve provides:
o a first slow learning phase
o a second fast learning phase
o a third slow-learning phase tending to an asymptote
The variants to the canonical model are the:
o multi-stage S-curves
o asymmetrical S curves
