COMMENTS , RESULTS AND ANALYSI S OF TH E OLD
SPINNIN G MILL S OF SAO PAULO
f.... • A. General comments
' 132. The individual productivity of the old spinning
mills of Sao Paulo included in the sample, is shown in
graph 1, together with the general averages and the productivity of the standard mills which served as a norm
of comparison.
.133. A general examination of the graph reveals great
Consistency among the results obtained in the old spinning mills* with one exception (no. 9), the productivity
of which is much higher than that of the others. Taking
into account the fact that these mills are of varying sizes
and degrees of modernity, the consistency of the results
may imply the presence of a general factor, independent
of size or type of equipment, which affects all the mills
more or less uniformly. The exception of mill no. 9,
however,, indicátés that this factor'is not completely beyond the control of mill managers. According to ob-'
servationsmade in the spinning mills of both Sao Paulo
and Rio dé Janeiro-Distrito Federal, this factor would'
appear to be the existence of antiquated working
methods. ' 


'
134. Mill no. 9 is equipped with modern European
machinery for opening and picking and for a part of
the card section, but the remainder dates from 1912 to
1930. It can be said that roughly a quarter of the mills
is modern. Even so, its productivity is higher than that
of mill no. 7, which is of approximately the same size
and has twice as much modern machinery as mill no. 9.
As both mills work under the same general management,
but are controlled by different superintendents, the difference in productivity may indicate that the degree of
modernization has not had a very important effect compared to other factors, and that probably the difference
of internal performance between the two mills can be
attributed to the effectiveness of direct supervision.
• 135. The relatively .small influence of the degree of
modernization is also apparent from the comparison of
mills nos. 8 and 7, for which the levels of productivity
are fairly similar, though the former is typically old in
all its processes, while as already stated, 50 per cent of
the latter's machinery is modern. If differences of size
are taken into account, the productivity of mill no. 7 (on
the extreme right of the graph) is relatively low, compared with that of mill no. 10, which is old though its
spinning frames are equipped with long-draft systems.
136. The type of cotton used does not appear to cause
important differences in productivity, since during the
course of the investigation mills of both high and low
productivity were found to be using the same type of
cotton. This does not exclude the possibility of the influence of all types of Brazilian cotton on the industry
as a whole.
137. With the exception of the left part of curve no. 7,
which represents a rather unusual example of coarse
combed spinning, the curves of the old spinning mills
appear ranged progressively according to size, both in
the section of the graph corresponding to carded yarn
and that of combed yarn. Their relative productivity,
however, does not vary according to the theoretical influence of size, since this influence is obscured by other
more important influences.
138. The following conditions were observed in most
of the mills:
(o) The quality of the intermediate products is defective ;
'.' (b) The cleaning of the air, machines and floors is
poor;
(c) The working condition of the machinery is only
fair; in many mills maintenance was found to be carelessly carried out; . 


••
(d) Control of quality is non-existent or very defective. Most of the mills have only the instruments which,
are indispensable for determining the weight and
strength of the yarn;
' (e) Control of efficiency is practically non-existent ;
(/.). The lighting is defective. In most mills natural
Chapter II. Brazil 21
Graph No. 1
PRODUCTIVITY OF TH E SPINNING MILLS INCLUDED IN TH E SAMPLE
Brazil (Sao Paulo)
10.01 — — — —
9.0 .
o.i L
O.. L
o,L
0.3Jo.aL
O 10 2 0 3 0 4 0 SO 6 0 7 0 6 0
X—Yarn count K—Means carded yarn
Y—Productivity in kilograms per man-hour C—Means combed yarn
I—Average of the old mills A—Means an old mill
II—Average of the modern mills M—Means a modern mill
III—Standard productivity of an old 25000-spindle mill The first number of a mill's key is a reference. The
IV—Standard productivity of a modern 25000-spindle mill number underneath is the size of the mill in spindles.
22 Productivity of the Cotton Textile Industry in Latin America
daylight from skylights is used, together with incandescent lamps ;
(g) Work-loads are not determined by rational
methods; -
(h) Minors are allowed to work, under the supervision of governmental agencies. In view of their inexperience and naturally restricted capacity, more tenders
are needed per machine or to carry out any other tasks;
(i) As a result of the industrial development of Sao
Paulo, there is a large percentage of labour turn-over in
the textile mills, and consequently, many enterprises are
unable to provide new workers with adequate training;
(/) In some mills defective methods of work were
observed, especially as regards doffing and yarn hauling;
(k) In the time available for visiting mills, it was not
possible to determine whether the type of cotton used in
Brazilian spinning mills did or did not affect productivity.60 In many mills the card web was found to be
neppy, even at normal doffer speeds; on the other hand,
cases were found where the card web was perfect with
three different types of Brazilian cotton. This indicates
that the method of processing the cotton, or the adjustments and speeds of the machines, have probably not
been adapted accurately for Brazilian cotton. Some manufacturers stated that Brazilian cotton led to excessive
waste at the pickers and the cards.
B. Analysis of the results
139. Table 3 is a summary of the average values of
the influences, that is, the indices of the importance of
the factors affecting productivity.61 The total influence
and its principal components—the influences of opera-,
tion, type of equipment and size—are the averages of
the influences of each yarn count, which were obtained
in table 4 by means of comparisons between the actual
and the standard consumption of labour per kilogramme
of yarn.
140. The influence of size (1.02) is of no importance,
as far as the mills of the sample are concerned. However,
the weighted average of the influences of size in all the
spinning mills of Brazil is 1.08, indicating that the sample was mainly made up of larger than average size
mills for Brazil. The choice was made bearing in mind
that in the distribution of the industry according to size
(table 5), the large mills of 15,000 spindles or over
represent almost two-thirds of the total capacity and that
general conclusions regarding productivity would have
been the same, had a greater number of small mills been
included.
°° The principal characteristics of Brazilian cottons are as
follows:
Paulista: Grades 2 to 7 (approx. GM to SGO) ; length of
staple 24/26 to 32/34 mm.;


 uniformity slightly better than average; character slightly better than average; colour white;
strength good; elasticity fair; cleanliness good; in the average
and high grades, the waste of pickers and cards is from 9 per
cent to 11 per cent, and in the lower grades it reaches 13 per
cent; card web tests show 19 to 30 neps per 100 sq. inches; the
fineness is from 3.7 to 4.8 microgrammes per inch of fibre (av.
of 4.2 to 4.4) ; the appearance of the yarn is usually better than
average; it is used mainly for coarse and medium yarns up to
30's; it is similar to the Texas and Express varieties from which
it originated.
Serido: grades 2 to 4 (approx. GM to M) ; length of staple
32/34 to 36/38 mm; uniformity average; colour white; strength
good; elasticity excellent; cleanliness fair (contains foreign
matter, neps and dead fibres) ; it is used for fine yarns from
SO's to 90's.
Sertao: grades 3 to 6 (approx. SM to SLM ) ; length of staple
30/32 to 32/34 mm; uniformity poor to average; colour ivory;
141. The influence of the type of equipment (1.34)
was calculated assuming that all the mills were entirely
old, as it would have been very laborious to prepare
standards for the degree of modernity of each mill. It
should therefore be remembered that this value is, in
fact, less than 1.34. The number of long draft and
standard draft spindles in each Brazilian state can be
found in table 6.
142. The influence of operation (1.96), the most important component of total influence, was broken down
into the influences of the draft schedule, of speed, of
efficiency, and of the excess of direct, indirect and miscellaneous labour, by means of an examination of the two
mills showing the highest and the lowest productivity
(tables 7 and 8). Averages were taken of the influences
of all the processes of these mills and, in table 17, they
were adjusted so that the result was equal to the influence of operation which had already been determined
by means of the general sample.
143. 


The redistribution of the influences was carried
out in the light of the following considerations, which
emerge from a study of the tables of analysis mentioned
above, and of the general observations made in the
industry.
144. Low speeds were found at the cards; the drawing frames and the spinning frames. At the cards, this
is generally due to (1) the poor condition of the card
clothings, which could certainly be corrected by adequate
maintenance, and (2) the need for giving the cotton a
more thorough cleaning, at lower doffer speeds; this
could not be achieved by any other means, unless the
opening and picking equipment is modernized. There
was no proof that the low speed of the drawing frames
is due to any outstanding cause and in any case its correction was considered simple. In some spinning frames,
speed has been reduced as a partial compensation for the
excessive wear on the machines, but it is estimated that,
on the whole, the spinning frames of Sao Paulo could
work at normal speeds if they were carefully maintained and if the quality of the product were improved
in the earlier processes. Thus, this difficulty can likewise
be remedied.
145. In view of these observations, it was decided to
divide the influence of speed (1-07) into two parts,
•namely, that which could be corrected (1.04) and another which could not (1.03).
146. The existence of defective manufacturing conditions, such as the excessive wear on machinery and lack
strength fair; elasticity good; cleanliness poor to average (contains foreign matter, neps and a great amount of dead fibre) ;
fibre silky; it is used for yarns from 30's to 50's.
Matas: grades 4 to 5 (approx. M to SLM) ; length of staple
22 to 26 mm.; uniformity poor; colour beige; strength less than
average; elasticity less than average; cleanliness poor; fibres
rough; it is used for coarse yarns from 6's to 20's.
81 If 1 is substracted from these influences and the result is
multiplied by 100, they become the actual percentages of the
excess labour per kilogramme produced; they also become the
potential increases of productivity, expressed as a percentage
of present productivity, which could be obtained by the elimination of the causes affecting it. The component or partial influences have the same meaning as the total influences, but they
refer to specific causes. When the influences of two or more
causes are multiplied together, the product is the influence of the
combination of these causes. The reciprocal of an influence,
subtracted from 1 and multiplied by 100, is the loss of productivity expressed as a percentage,


 arising from the cause corresponding to that influence.
Chapter II. Brazil 23
of maintenance, lighting and cleaning, would normally
be conducive to low efficiency in all the processes, especially in the spinning frames. Since this is not the case, as
can be seen from the influences of efficiency in both mills,
which are almost normal or even better than normal, this
condition indicates that part of the total excess labour is
employed in counteracting these manufacturing conditions and raising efficiency. In order to estimate a value
for this part of the excess and separate it from what has
been termed absolutely superfluous personnel, the breakage frequency of yarn was measured in the spinning
frame section of several mills. The conclusion was drawn
that manufacturing conditions probably require IS to
30 per cent more personnel than indicated by the standard work-loads. In order to simplify the analysis, an
excess of 23 per cent, that is, an influence of 1.23, has
been ascribed to defective manufacturing conditions, as
a whole, with the exception of low speeds. As the influence of efficiency represents 1.05, the remainder, 1.17,
must be separated from the influence of the total labour
excess (1.75) in order to determine, by division, the
influence of the absolutely superfluous personnel (1.50),
that is, the proportion which can be eliminated without
correcting manufacturing conditions. An inspection of
conditions in the mill indicated that both the influence of
efficiency (1.05) and the fraction of excess labour influence corresponding to manufacturing conditions (1.17)
may be divided equally between the causes which can be
corrected and those which cannot; this gives rise to the
figures 1.02, 1.03, 1.08 and 1.09 which appear in the
summary of the analysis.
147. As can be seen in the tables of analysis of mills
"A" and "B", a superfluity of personnel occurs in practically all sections, although the effect on the general productivity is more important in the spinning frame section
because its consumption of labour per kilogramme is
normally quite high. Since the influence of direct labour
is very high and that of indirect labour very low, even
lower than 1, it is evident that labour is wasted not only
because few machine units are assigned per tender, but
because the relative lack of indirect labour forces operators to carry out auxiliary tasks which they should not
have to fulfill. Specialization of work alone would probably enable the industry to make considerable reductions
in personnel, without demanding of the tenders a greater
effort than they are making at present.
148. The improvement of working methods, especially as regards transport and doffing operations, could
also bring about an immediate reduction of personnel.
Reference has already been made to the fact that in
Brazilian mills, frequent cases of manual transport were
encountered when containers or small wheeled-carriers
might have been used.
149. In the lower part of the analytical summary table,
the influences have been rearranged in groups corresponding to: (1) causes which require only immediate
action by the management to reduce the personnel; (2)
causes which call for preliminary action by the management, in order to improve certain manufacturing conditions ; (3) causes which can be corrected without modernization ; and (4) causes which can only be eliminated
by means of the modernization of the industry.
150. The increase in productivity which could be
achieved by means of the reduction of personnel and the
improvement of manufacturing conditions, demands the
establishment of adequate systems for determining
work-loads and controlling efficiency and the quality of
the products. It would also be necessary to select the
best workers and to train them so that they can'adjust
themselves to the new conditions.
III . COMMENTS , RESULTS AND ANALYSIS OF TH E
MODERN SPINNING MILLS OF SAO PAULO
A. General comments
151. The individual productivity of the modern spinning mills of Sao Paulo included in the sample is shown
in graph 1, together with the general averages and the
productivity of the standard mills which were adopted
as norms for comparison.
152. General observation of the chart reveals that
good productivity conditions prevail in the modern spinning industry of this region. If the influence of the size
of the mills is taken into account, the observations reflect
conditions varying from 54 to 103 per cent of the productivities set up as standard.
153. All the mills manufacture a limited number of
yarns and are outstanding for the high quality of their
management. The productivity of mills nos. 3 and 5
should be higher since conditions of manufacturing and
supervision there are excellent. However, they were designed to make very fine yarns and, at the time of the
investigation, market conditions obliged them to manufacture yarns of a much lower count. 


As a result, their
processes were thrown off balance, causing a reduction
of productivity.
154. There is no correlation between the size of the
units and their productivity, because surplus labour is a
more important factor than differences of size. Nor does
the chart indicate any differences of productivity caused
by the type of cotton used. On the contrary, the modern
spinning mills of Sao Paulo prove that with good equipment and adequate technical management, certain defects in the cotton have no appreciable effect on productivity, although they may have on total raw material
costs.
155. On the whole, conditions in the mills are fairly
good, except that there are more workers than necessary.
Owing to the low cost of wages, in relation to the cost
of investment in the mills, attempts are made to obtain
high efficiency rates in the processes at the expense of
labour productivity. The systems of quality control
adopted for the intermediate and final products are
good; most of the mills are equipped with laboratories
and well-trained technicians. On the other hand, not
much attention is given to the rational determination of
work-loads, probably because there is little incentive to
assign a maximum number of machines per man.