## There first approach is Winkler model which

The building frames are designed as 3D space frame consisting of four node elements with appropriate dimensions by using SAP2000 software. Slabs and mat foundations are modeled as rigid diaphragm with the help of four node plate elements. The raft foundation is used support structure and it is designed by conventional method assuming that the foundation to be rigid. In modeling the 3D elastic continuum model (FEM model), the finite element method is used. The steps of the finite element methods are a. Pre-processing that includes mesh generation, b. Obtaining the assembled system of equation, for which the elemental matrices and vectors need to be evaluated, c. Applying the boundary conditions, d. Solving the linear system of equations, e. Post-processing. For analysis, the finite element is adopted in this study because of its diversity and flexibility as an analysis tool. The soil- structure interaction (SSI) is represented by using equivalent springs with six degrees of freedom. In two horizontal and vertical axes with rotational springs about those mutually perpendicular axes are considered. These are attached to estimate the effect of soil flexibility. George Gazetas formula is used for determining the stiffness along the six degrees of freedom. The time history analysis is adopted for RC frame structure. Therefore, the earth quack ground motion (Bhuj EQ data) is subjected to the whole system of RC frame structure. According to the results of the analysis, it can be seen that lateral displacement increases as the number of story increases. The increase in the stiffness of soil makes a reduction in displacement. Raft thickness does not have any effects on the displacement of the building. Further, it has been observed that the time period of the building on a different type of soils increases more than the time period in the fixed condition. For instance, the maximum time periods ae observed in FEM models, while the minimum time periods are found in fixed models. When the time period increases, the natural frequency of the building decreases. Finally, as the soil flexibility increases, the bending moment also increases with the increase of the raft thickness and height of the building.

## Innovation speed as well. The design team

Innovation in Zara

Zara innovates on various fronts and its innovation can be categorised into various categories. The two most important innovation categories of Zara are speed and presentation.

Speed

Zara is in the fast fashion industry, in which product life is short and differentiation is crucial to build a strong image of the brand. The most important contribution of Zara is process innovation, fast fashion, and not a new product. Zara’s production locations and shipping methods are both rather expensive in order to be able to be fast at reacting to trends. This part of Zara’s business model has been put into use, when the founder of Zara, Amancia Ortega, figured out that a big part of the key of dominance consisted of speed and responsiveness. Furthermore, Zara innovates on the production speed as well. The design team of Zara has the job of generating 18,000 new product designs a year, which is equal to more than 70 every working day. The design team has 350 people available for this task. Loreto Garcia, a designer explains, ‘We get our ideas from books, magazines, twitter and blogs but above all from feedback from our shops. Our customers tell us what they like and don’t like. We attend shows and talks about fashion to get ideas. It is a continuous process.’ The communication isn’t limited between customers and the company, staff of the stores are making use of wireless communications with the headquarters to send data about sales and inventories. This fast way of communication enables Zara to work with the buying patterns of the customers on a short term, therefore Zara has the ability of supplying the most wanted designs everywhere needed.

Another innovation of Zara regarding speed is related to a new idea to minimize the risk of oversupply. Zara starts off with a relatively low pre-season inventory commitment, which consequences in a higher competence of reacting quickly to new trends and orders. The entire process of a new product, from the initial idea till the moment on the shelf, takes only four weeks and Zara keeps most items for less than two weeks in stores as Zara has to innovate. Process innovation appears to be a very good innovation for Zara as product innovation can be imitated in a very short period of time, while the process innovation model continues to be profitable.

Presentation products

Zara had an idea to better present the outfits and encourage customers to buy combined outfits: displaying the garments together instead of on the regular way, everything sorted by category. Next to that, Zara owns 25 full-size shop windows with differing displays and lighting to display the image of the stores under different circumstances to the designers. This method innovates by enabling designers to test designs and ideas quickly and thoroughly, resulting in good image of the stores and thereby Zara.

Joint-ventures

Zara had learned from other global retailers that had often failed when expanding to foreign countries. As a result, Zara became extra cautious and started with joint ventures when expanding to foreign countries to gain access to prime real-estate and local expertise and to reduce investment risks. The next step in their strategy is buying partners out within multiple years as Mr. Jose M. Alvarez says: “We don’t mind investing in joint ventures to learn, but prefer to go alone”.

Perform an analysis of how Zara innovates.

Explain what constitutes ‘innovation’ in the context of Zara.

In the book ‘Exploring Strategy’ by Johnson et al. (2017) 4 innovation dilemmas are described. Explain how Zara deals with two of the innovation dilemmas: (i) The dilemma of Open or Closed innovation, and (ii) The dilemma of Product or Process innovation.

Entrepreneurs innovate by identifying and exploiting new ideas and inventions that result in innovations.

Innovation involves the conversion of new knowledge into a new product, process or service and putting of this new product, process or service into actual commercial use.

## Supervised no dependency on the parameter ‘f’.

Supervised
Learning Coursework 2, GI01/M055, 2017/2018

Solution to
Question 1 :

(a)

·
A prior distribution – probability distribution over a function dp(f)

·
An additive noise model – inaccuracy in the output observations.

The Noise for each measure assumed to be independent.

(b)

According
to the Bayes theorem – Posterior probability of a parameters (f) given an input
data (X) is the product of the Likelihood function and Prior probability dP(f)  (prior knowledge before performing the
experiment) divide by probability of input data.

Let
us assume the inputs data x = {Xi, where i = 1,2,3…..N} and its
corresponding  random function of f  = { fi, where i = 1,2,3…..N }

dPpost(f|X,
y) = (dP(f) P(y|X, f))/p(y|X)

We
can eliminate the denominator p(y|X) since there is no dependency on the
parameter ‘f’.

Therefore:

dPpost(f|X,
y) = (dP(f) P(y|X, f))/p(y|X)  ? dP(f) P(y|X, f)

dPpost(f)
= dP(f) P(y|X, f))

(c)

We
know that in the case of Gaussian process regression the prior and additive
noise model are Gaussian.

Let
=
, where

P(y|X,
f)) =   ,
where X Î  and y Î

The
data corrupted by likelihood function with zero mean and some variance

Prior
is given by dP(f) for symmetrical Gaussian distribution – centred at the origin
over the weight vector w.

(d)

For
a finite collections of random variable stochastic Gaussian process has
multivariate normal distribution while for infinite random variable it has a
joint distribution.  In our case, it is a
1-dimensional Gaussian(normal) distribution know as marginal distribution.

(e)

Marginal
distribution is the integral of prior multiplied by additive noise model.

Let
us assume the inputs data x = {Xi, where i = 1,2,3…..N}

its
corresponding random function of f = { fi, where i = 1,2,3…..N } and
output v    alues yi …… yn

P(y|X,
f)) =

The
posterior distribution:

By
manipulating the above expression, we can get the mean and covariance matrix.

The
argmin of the weight vector( ) would maximise the posterior i.e it would
give us the mean of the distribution.

is the mean distribution

,

where  is standard optimiser,

is the identity matrix which is

And the covariance  can be expressed as:

(f)

The
full Bayesian inference adds up all the posterior distribution output and then
takes the average as its output. The maximum a posteriori probability (MAP) picks
a weight vector which maximises the posterior distribution.  If the prior and additive noise model are
Gaussian then the Posterior is also a Gaussian. In linear Gaussian, the
convergence of the posterior mean and the MAP estimator coincide.  The sample mean of random linear function is equivalent
to the MAP mean of Gaussian for any value of standard deviation.

Since
we have prior knowledge about pervious posterior distribution, we can use it to
enhance the MAP hence we can get the complete posterior distribution – as shown
above in part ‘e’ the Gaussian have a mean of  in high dimension which is more accurate than sample mean
from the linear function.

Solution to
Question 2 :

(a)
When the value of a =b =1, the Beta distribution
is equivalent to uniform distribution. If we integrate B(a,b) and compute in interval of
0,1 we get 0.5 mean and a /a +b gives 0.5 mean as well.

(b)

Let X be the data X =
{1,….n}

H ={h:i?X ?0,1}=0,1X

Prior given as, P(h) =

and noise model P(y = 1|h,(i,y)) = h(i) and P(y = 0|h,(i,y)) = 1?
h(i)

Posterior distribution is
the proportional to the product of prior and noise model.

Prior of the beta
distribution is given by

P(h) =

Ppost =

,

This means
Beta is conjugate to the Bernoulli.

(c)

The mode is equivalent to the
maximum a posterior distribution, so

Mode of the posterior

The maximum
likelihood estimate is the empirical probability, i.e. ,

Mean of the
posterior E(p|X) =

For uniform distribution
where a =b =1

P(y = 1|h,(i,y)) = h(i) = 0.3

P(y = 0|h,(i,y)) = 1-h(i) = 0.7

Therefore,

Mean = 0.4333

Mode = = 0.3

Solution to
Question 3 :

(a)

Overfitting or High variance is when the learned hypothesis fits
training data well but fails to generalize for unseen dataset. The model might
be too wiggly which is not a good model to predict future(test) dataset. In
other words, the predictor is too complex and not consistent for unseen data.

(b)

Empirical risk minimization(ERM) – the learner algorithm gets X training
data input and model function H which has an output predictor H(x) from unknown
distribution D. The goal of the learner algorithm is to minimize hypothesis H*
for fixed function H for which E(L(H(x), y)) is minimum.  For such unknown distribution, we minimize
the empirical risk over the given training dataset such technique in statistical
learning algorithm known as Empirical risk minimization(ERM.)

(c)

Structural Risk Minimization(SRM) is on finite dataset, the task
of selecting best model class function within a nested family of space .  The model may be too complex at HQ  and poor at generalizing it to a new
dataset which would consequent an overfitting problem. The SRM addresses
overfitting problem by solving such problem by balancing the model’s complexity
and training data error.

Let  in Hq minimizer be

As we increase q, the   gets reduced, the   function increase in between we get a
balanced model.

(d)

Regularization addresses overfitting problem by penalizing complex
learning algorithm – it selects one complex hypothesis class and adds regularizer function  to empirical error to prevent overfitting.

Our goal is to choose best learning algorithm among many possible
hypothesis space – the regularization parameter can be chosen best   in case of ridge regration or best  in .

If we have large dataset – we can split the dataset into three
different parts:

Training dataset – where we model our functions using different
learning algorithms.

Development or Validation dataset – on this we tune the parameters
on held out dataset (development or validation set.)

Test
dataset – to test best model we have developed on the training dataset.

(e)

The kernel trick enables us to work with linear function in
infinite dimensional feature space.

subject to yi?w,xi??1??i,?i ?0, ?i = 1,…,m.??

Hinge loss function is defined as:

Basically, we minimize the hinge loss while minimizing the norm of
the weight vector(w) that corresponds to maximising the margin. Maximising the
margin gives a good generalization learning algorithm even though we have a
high dimensional feature space. The nonnegative xi(?i) measures the hinge loss – computes
the amount by which fails to meet a margin of 1.

(f)

By changing the hinge loss function, we can create a linear
programming boosting.   The primal

subject to yiHia?p??i,?i ?0, ?
= 1,…,m.?

and the dual would have the
following form:

;

subject to

where  is the distribution,

Optimal solution can
be achieved by doing the dual programming iteratively. We assume the set of weak
learners are finite. The boosting algorithm would give optimum error bound.

We might have more than one weak learner’s optimal solution which
are approximate to the cost function.

Solution to
Question 4 :

(a)
In statistical learning theory generalisation error is a measure how
well a predictive of a learning algorithm perform on unseen dataset.  The expected error is approximately equivalent
to the empirical error and the error reduces when we increase the dataset. Since
the generalisation of the learning algorithm is done using randomly drawn from
a finite independent and identically distributed samples the model might be
sensitive. One way of overcoming this problem is by relating the
training(given) error to the generalisation error – which possibly would avoid
overfitting.

There are many theories for
generalisation error – each theory has strengths and weaknesses.

(b)

For a fixed size of ‘m’
training datasets

None linearity in the lost
function would give a small difference in the average of the distribution.

Analysing the mean using

,

If we use consistency of
classification method for big dataset

where   if p(x,1)>p(x,0), otherwise

This might not be a good solution when we have small sample size.

We are taking limit m tends to infinity so for small m the Bayes risk might
fail.

The 95th percentile is 0.15 which is above the mean value 0.08.

This means the probability we have been misled by 0.05 which is less than the
mean value.

(c)

Structural risk minimization(SRM) is a best learning algorithm selection
procedure – choosing the model that minimizes VC upper bound on risk.

,

For fixed datasets, as the complexity of the learning algorithm C(Hi)
increases, the training error (eˆrr(c)) will decrease but fails to generalize for unseen dataset.  On the other hand, as the VC dimension
increases,   will increase since this depends on the VC
dimension. SRC chooses a best model from the sequence of hypothesis ( which minimizes the right-hand side of the
above inequality.

As the complex models might over fit and at the same time the more
training set we have the lower empirical error will be. SRC resolves the trade-off
by providing a measure of characterization between complexity and empirical
error – by
balancing the model’s complexity and training set (empirical) error.

(d)

As discussed above(b)  is the probability that the training data misled.

Assuming
Hk is finite and confidence at least 1-, where  is small positive number, we can compute the likelihood
of being misled by evaluating training error equals to zero. In our case,

Since the functions are drawn
randomly from a distribution we must add prior weight. By applying Hoeffding’s inequality
to the randomly drawn probability we can get the generalization error is
bounded – for model k.

Reference:

I have used the lecture notes
for this assignment.

https://moodle.ucl.ac.uk/course/view.php?id=11543

## Introduction new ways to design and make

Introduction

I have decided to have an emphasis on corsetry and lingerie in my Fashion practice. I’ve always had a keen interest in how the female form has been portrayed in art throughout history, however; I have now changed my perception to see how women choose to portray themselves and how they achieve the desired look expected of them by society. Corsetry dates back as far as 1600 BC and has been used since as part of a woman’s daily routine to either emphasise and exaggerate her assets (Polaire, 1874-1939) or to create the illusion of a very boyish and straight figure (1920s). Although corsets have been used and designed for centuries, this doesn’t deter me from being inquisitive in finding new ways to design and make a wearable and contemporary garment.

Figure 1. Saunders, L, (2017), Draping on the stand, Calico Cotton, UCA

During Stage 1 of the course I made a corset like garment which incorporated three dimensional textures to add an aesthetically pleasing design so it could be worn as an outer garment.  The texture was predominantly around the bust area which added more volume to the upper part creating a fuller breast whereas the waist is cinched, creating visually a smaller waist despite it not having the boned structure to actually do so. My design is a contemporary response to a traditional corset however, it does resemble more so a bustier. Although it is a contemporary design, the ruffled texture has a historical link back to Victorian dresses which incorporated this design to add volume and aesthetics.

Figure 2. Saunders, L, (2017) Fashion Illustration for Mementos Project, black pen on paper, dimensions, UCA

My second piece of work is an illustration for one of my final piece ideas. It incorporates the traditional corset aspect of creating a smaller waist and fuller bust however it is inspired by 1920’s art deco. Although the fashion in the 1920’s was to have a very straight and boyish figure, i gathered that the combination of the culture of the decade and a traditional corset garment would come together to create a very modern design. The linear designs would also cleverly act in accentuating the waistline and breasts. The materials I use would reflect that of the 1920’s such as old gold satin and black jewels/sequins for the embroidery.

Pre-Modernism (before 1850)

Figure 3, Unknown, (16th Centaury), Silk and Linen, German Bodies

‘The German pair of bodies buried with Pfaltzgrafin Dorothea Sabine von Neuberg in 1598’ is the only known corset which dates back to the 16th century (Figure 3, Arnold, J., Tiramani, J. and Levey, S., 2008, Patterns of Fashion, p.47).

Made up of three layers including silk and linen, a pocket was incorporated at the front of the garment so that a stiff busk (a rectangular piece of ivory, wood, or whalebone) could be inserted. Ranging from 10 to 14 inches in length, and 3/8 of an inch thick, the busk flattened the torso and breast area (pushing the breasts up) creating the desired look of the period. Busks are not always necessary if the wearer has a small chest and the corset is already heavily boned.  Due to its use as an undergarment, the dull cream colour helps to keep it discreet. Tabs and eyelets at the waist helped the wearer to secure and attach the farthingale (stiffened hoop skirt) or petticoat. Although seemingly ingenious, this technology adapted the corset into a practical garment. As seen in Figure 3, the arm holes are quite far back. This was common in the 16th century as it forced a stiff and rigid posture which was considered ‘good breeding’.

A modern audience viewing this garment may see it as quite extreme. However, despite traditional methods being altered (such as the busk), corsetry throughout the years and to this day remain to achieve a similar goal of a flat torso and a smaller waist. Although not intentional, but this garment has a feminist perspective and is an example of how for many centuries woman have to contour their bodies into unnatural shapes to please the male and society’s expectation. This work has influenced my outlook on my past and future projects as I aim to create a modern day corset for the woman but not for her to feel as though she has to alter her body for the satisfaction of others.

Contemporary (1970 – present)

Figure 5, Jean-Paul Gaultier, (2012), Metal, Fall 2012 Collection Gold Cage Corset

A contemporary, almost futuristic garment designed my French designer Jean-Paul Gaultier. Although deemed contemporary, I can’t help but see inspiration and influence form 1920’s art deco. The linear construction and vibrant gold colouring is very typical of that time period. An excellent example of how a designer can subconsciously be influenced by precious works of art or time period and incorporate it into their garments to create a fresh outlook on a contemporary design. The use of the metal resembles 16th century orthopaedic corsets, from a modern perspective these corsets will be viewed as extreme and unpleasant however, when constructed in an aesthetically pleasing way it isn’t seen in the same light. The corset itself is abstract as it’s not your conventional garment and wouldn’t be used in the same way a traditional corset would. It is designed as an outer garment due to its incredible detail and rich colouring. This design to me means more of a celebration of the female form rather than degrading the woman to have to change her body. The large bust and hip area emphasises her natural shape and the gold celebrates and frames her body.

Again, I see this design as a feministic piece as it is celebrating the natural form of the model and emphasising her natural features. This design has shown me that there are ways to create a garment that was once used to suppress women and turn it into something shows off a woman natural form.Figure 2. Saunders, L, (2017) Fashion Illustration for Mementos Project, black pen on paper, dimensions, UCA

My second piece of work is an illustration for one of my final piece ideas. It incorporates the traditional corset aspect of creating a smaller waist and fuller bust however it is inspired by 1920’s art deco. Although the fashion in the 1920’s was to have a very straight and boyish figure, i gathered that the combination of the culture of the decade and a traditional corset garment would come together to create a very modern design. The linear designs would also cleverly act in accentuating the waistline and breasts. The materials I use would reflect that of the 1920’s such as old gold satin and black jewels/sequins for the embroidery.

## Syria Kurds fighting for the SDF to

Syria is no
stranger to conflict. The Syrian Civil War, a multi sided conflict between the
government of President Bashar al-Assad, with its allies, and various opposing
forces, has persisted approximately six years. The U.S. has recently begun plans to
create an allied Syrian Democratic Forces (SDF), Kurdish-led, militia border
security force.OM1  The U.S. intended to use the back bone of this coalition borderOM2  force, known as the Kurdish People’s Protection Unit (YPG), in its fight
against the Islamic State (IS). However, Turkey OM3 views this strategy as counter productive and detrimental to not only
their national security but to U.S./ Turkey relations. President Erdogan of
Turkey considers the Kurds fighting for the SDF to be a terrorist organization and
has since condemned the SDF, YPG, and U.S. commingling. The YPG is understood
to be an active extension of the banned Kurdistan Workers’ Party (PKK), known
for fighting for Kurdish autonomy in Turkey for thirty years. “Turkey’s
president vowed to “suffocate” efforts to begin training members of
the Syrian Democratic Forces (SDF) and create what he called a “terror armyOM4 “. The coalitions goal is to create a force of approximately 30,000
fighters over several years of which the Kurdish forces and Arab SDF will each
make up half. The BordeOM5 r Security Force (BSF) will be tasked with patrolling large sections of
Syria’s northern border with Turkey and the eastern border with Iraq. The U.S.
led coalition hopes that creating a strong border OM6 force will help restrict Islamic State movement and deter the
transportation of illegal substancesOM7 .

OM1Is
the SDF a “Kurdish-led, militia border etc.?” If so, separate the title and
definition with a colon or “a.” Otherwise, it seems like a list of two
different things.

OM2border

OM3run
on sentence

OM4

OM5border

OM6border

OM7

## The and expanded around the world; this

The historical act of the world
transcendence was carried out with the Independence of the thirteen Colonies on
July 4, 1776, according to which the Colonies achieved independence signing the
declaration of independence drafted by Thomas Jefferson, the one that establishes
a true philosophy of freedom. In this declaration, it is held that all men are
born equally by God’s work that deliberates fundamental rights among which are
life, liberty, and the pursuit of happiness. According to the article The Declaration of Independence: The Words
Heard Around the World, the independence of the United States caused the
ideas of freedom to be expressed and expanded around the world; this historical
fact had universal repercussion since it was not only felt in the French Revolution,
but it was also a living example for the Emancipation of the Colonies of South
America, such as: Argentina, Chile, New Granada, Venezuela, Bolivia, Quito, and
Peru. The French Revolution, in 1789, took the same ideal of freedom to the
detriment of the monarchy, in addition to the ideals of equality and fraternity
among men, inspired by the philosophical and intellectual movement of the
Enlightenment.

The impact was created because, for the first
time, a colony was emancipated and demonstrated the success of the Republican
and Constitutional Government. As we can see, history shows us how the act of
colonies of Central and South America of the Spanish oppression. The
independence of the United States showed that it was possible for the colonized
territories to be liberated from the European oppression. The American nation
was a point of reference for many Latin American leaders. After achieving
independence, many of the former Spanish colonies were inspired by the
political organization of the United States to build their states.

## Any is a stressed situation and they

Any of the officers
were not allowed to leave Vietnam
unless they clear the drug addiction. Lee Robins was the main the psychiatric researcher. Her studies showed that when
soldiers came back to U.S. 95 percent of them were clean and without addiction.

The normal
drug abuser shows the following routine- they get clean from hospitals or rehab
centers but as soon as they go to their
home they get re-addicted but Vietnam soldiers were different. The habits were
triggered in certain situations and under certain conditions.

How did all these things started in
war?

The war is a stressed situation and they were in contact
with fellow soldiers who were consuming heroin. One of the influencing factors
that lead to formation of any behavior is surrounding environment. For example-
if you’re the only person who doesn’t consume
heroin and all your surrounding people are consuming in front of you, then they
start forcing you to just try a little
bit. But ones you’re involved then it’s getting difficult
to get away from it, especially when you are in stressed situation. We get
pushed away in situations.

But once
stress and where there is no fellow heroin user
is the good surrounding for him. In war, there were no family and no one to
stop them from using heroin. But at home,
you are surrounded by your loving family, friends and you don’t think about
consuming anything. So the absence of any
external stimuli also triggered the addiction.

The most important
lesson we can learn from soldiers is you can change your behavior or control yourself in the influence of external stimuli.

·
Don’t take or make any decisions under stressed situations.

·
If your friend or any relative is addicted to any drug abuse,
then be supportive and help them to get out of it. Motivation, emotions is
necessary to recover from drug abuse.

·
War is a stressed situation; it’s not easy to live alone
without family, many soldiers return from war with post-traumatic conditions.

·
The environment is the important factor, for example, if you want to learn something new then
surround yourself with the things related to that.

## Is in my ceiling? 1) The first

Is there a hole in your ceiling? And
looking for the solution on how to fix a hole in the ceiling? then this expert
anyone who has basic knowledge of handling the hardware tools. You can save a
ton of money on labour cost by doing it yourself. You require few hardware
tools which generally you can find in a nearby hardware store and some material
to fix it.

How do I fix a large hole in my ceiling?

1) The first step understanding the area.
Go to the area and see if there is any underlying cable around the ceiling hole
or electric wire underneath the hole. The wire could create a problem while you
fixing it. So first examine the area properly.

2) The next step is measuring the hole with
the measurement tool. Measure the depth and diameter of the existing hole. You
have to note down the measurement on the paper so you will use it later to find
the patch for the hole.

3) Now you need to find the right kind of
wood to patch the hole. Check your garage or any other wooden equipment which
is not in use. You have to find the wood which is the diameter of the hole to
use it as a patch. It will work as filler piece for the hole.

4) Once you find the right kind of wood for
filler. The next step is cutting it by using your taken measurement. Keep the
wood little thinner to provide a place for the drywall.

5) Now attach the wood to the hole with the
help of screws to patch it in the hole. Align the piece of the wood properly to
the ceiling.

6) Fill the patch with the coat of drywall
compound. Apply it gently all over the ceiling and give a smooth layer to the
drywall to look it professional. Once done, allow the compound to dry for 24 to
48 hours. Again apply the second coat of drywall compound and then paint it.

How do I repair a plasterboard ceiling?

You can use the same process that we have
used in patching the ceiling hole in the above guide. You have to find the
right kind of patch for plasterboard ceiling.

1) In the first step, we have to screw the
plaster washer to the ceiling all over the cracked area. While doing this you
have to make sure that you are driving screws into the wood lath of the
ceiling.

2) The next step is spreading the joint
compound onto the ceiling to cover the damaged area. Spread evenly all over the
area with the help of trowel.

3) Now put the insect screen into the
compound to give it a strong base to hold on. Again apply one more layer of the
compound and spread all over it.

4) Let the joint compound dry for at least
24 to 48 hours.

5) Once it is completely dry use the
sandpaper to remove an extra spot from the ceiling and make the ceiling look
smooth.

6) Now in the final step, you have to apply
the one coat of the primer followed by paint.

You have learned the ceiling repair
process. Now you are completely trained in the process. Get the right kind of
tool and start fixing it.

Go Top
x

Hi!
I'm Rita!

Would you like to get a custom essay? How about receiving a customized one?

Check it out