Flexibility

Loading...

Del curso dictado por University of Pennsylvania

Introducción a la Gestión de Operaciones

38 calificaciones

University of Pennsylvania

Introducción a la Gestión de Operaciones

38 calificaciones

Curso 3 de 5 en SpecializationFundamentos Empresariales

Aprende a analizar y a mejorar los procesos empresariales en servicios o en la industria. Aprende cómo incrementar la productividad y a ofrecer niveles de calidad mayores. Después de que finalices con éxito este curso, podrás aplicar estas habilidades en un reto empresarial del mundo real como parte de la Programa Especializado de Fundamentos Empresariales de Wharton.

De la lección

Calidad

La calidad no solo es la piedra angular de la gestión de operaciones, sino que también es un componente fundamental de la viabilidad y del éxito de una compañía a largo plazo. En este módulo se tratan todos los aspectos más importantes de la calidad en las operaciones, desde examinar las razones más comunes de los defectos, a las técnicas de implementación más usadas para abordar problemas de calidad y asegurar la fiabilidad y los estándares. Al final de este módulo, serás capaz de anticipar dónde podrán ocurrir los defectos y recomendar una estrategia sólida para mantener la calidad y la estabilidad.

Razones de Variedad6:10

Cálculos de capacidad con configuraciones10:42

Producción Modelo Mixto14:31

Elección de tamaño de un lote8:49

La fragmentación de la demanda y Fondo Común de Reserva9:31

Flexibility9:57

Plataformas y modularidad9:22

Límites a la elección del cliente5:17

Revisión del Módulo Variedad12:18

Conoce a los instructores

Christian Terwiesch
Andrew M. Heller Professor at the Wharton School, Senior Fellow Leonard Davis Institute for Health Economics Co-Director, Mack Institute of Innovation Management
The Wharton School

Increased product variety leads to more fragmented demand, more demand

uncertainty, and ultimately, an increased mismatch between supply and demand.

The major weapon against this mismatch is an increase in flexibility,

to the extent that we can produce exactly what demand requires,

there is no supply demand mismatch.

In this session,

we'll introduce a concept of shortening set-up times to increase flexibility.

We'll also talk about how you can configure a network of plants,

so that the plant network as a whole

is able to absorb the demand uncertainty benefiting from the effects of pooling.

First consider the effects of set-ups.

We spent a fair bit of time early on in this module to talk about

how set-ups reduce capacity.

To avoid this, we like to run long production runs which in turn,

however, leads to inventory.

The idea behind SMED is that rather than taking the set-up times as

a given exogenous variable, we start challenging the set-up time itself.

So why does it take 30 minutes to set up the machine?

What is actually going on during those 30 minutes?

The word SMED stands for a Single-Minute Exchange of Dies.

It's used to take automotive companies a couple of hours

to change between one state of production tools to another.

Typically, these are big stamping tools that weigh many, many tons, and

again, take hours and hours to change over.

SMED, single-minute, means we want to get from hours to minutes.

This, in turn, allows us to change the model much more frequently,

reducing inventory.

The key idea behind SMED is that

every set-up can be broken up into two types of activities.

They're called external activities and internal activities.

Think about the process of making the set-ups in our shirt example.

We noticed that some of the shirt set-up

included programming a CNC machine with a measurement of our particular customer.

Instead of doing this while the machine is in set-up, this could potentially be done

while the machine is still producing, cutting, another shirt type.

This allows us to change external set-ups that can be done

while the machine is running, and move them up front.

This cuts down on the set-up time that the machine is really standing still.

By then subsequently, improving the things that need to happen as part of external

set-ups and as part of internal set-ups, I reduce the total amount of set-up time.

Again, shorter set-ups now, mean that I can change models more often,

which is at the heart of mixed-model production, or

what we earlier on labeled as heijunka.

The idea of set-up time reduction at the idea of separating between internal and

external set-ups are by no means limited to production equipment.

Consider the following two examples.

Consider an airplane, the airplanes make money when they're in the air and

so the process of being on the ground, taxiing and

landing, standing at the gate is eating into their capacity.

What can we do based on the ideas of internal and external set-ups?

The idea of external set-ups is to try to shorten the time that the plane is at

the gate by doing some things ahead of time or potentially in parallel.

For example, one of the things that you know Southwest's doing

is carefully lining up the passengers while they are still at the terminal,

while the previous passengers are de-boarding.

This allows them to basically cut the time that it would take

new passengers to get on the airplane.

Similarly, think about cleaning the plane.

A lot of the activities related to cleaning the plane,

refilling it with drinks and meals, refueling it, is happening while,

for example, the passengers are de-boarding.

Similarly, we think about an operating room.

Now arguably, an operating room is not running big production batches of

ten patients with knee surgery in a row without set-ups but

an operating rooms needs to be set up between every procedure.

Again, think about the concept of internal and external set-ups.

Clearly, cleaning the operating room is an internal set-up,

this can only happen when the operating room is idle.

On the other hand, a lot of the patient preparation and

the anesthesiology work can happen on a patient outside the operating room

while the previous procedure is still going on.

Once the room is empty and clean, the patient on anesthesiology

can just simply be wheeled in and the overall set-up time is much shorter.

Consider a car company that offers ten models

that are made in ten different plants.

Each model is made in exactly one plant.

So you have model 1 that is made in plant 1, and

you have model 2 that is made in plant 2, and so on.

There's no flexibility in this plant network to have

one plant make multiple cars, or have one car being made in multiple plants.

Now, this is problematic as we saw earlier on in our session on pooling and

demand fragmentation.

This leads to a lot of variability in each of the plants or

put differently, if we could somehow pool the demand across models or

across plants, we would reduce the supply/demand mismatch.

Ideally, what we would love is we would like to have a plant network

where every plant would be able to make every one of the products.

This is the most flexibility that we can ever hope for, and

we would get the full benefits of pooling that we discussed in the earlier session.

Now the downside of such fully flexible plan at work is that you're going to have

lots of set-ups at each of the production settings.

If you take a look here at, for example, plant number 5, you're going to produce

many different products in this plant requiring many set-ups.

Moreover, you are going to invest a lot of money into tooling the plant so

that it is able to make each of the ten products.

So full flexibility is typically not a practical option.

As an alternative to full flexibility,

consider the concept, of what's known as partial flexibility.

The idea of partial flexibility is that you design the plan network so

that you assign every product to two plants and vice versa, you make every

one of the plants sufficiently flexible to produce at least two different products.

It can be shown that such partial flexibility is getting you almost

all the benefits of full flexibility but at dramatically lower cost.

Then we illustrate the concept of partial flexibility

in the case of the automotive industry.

On the left of the slide,

you see the plant vehicle assignment of Ford in the United States.

Here you see the plants and over here, you see the various product platforms.

Notice how the fourth plant network has relatively little flexibility.

In most cases, you have a vehicle dedicated to a plant, and

a plant dedicated to a vehicle.

On the right here, you see a nice application of partial flexibility.

Nissan's assignment from vehicles to plant

is really very much in the spirit of what I showed you on the previous slide.

Typically, plants are able to make multiple vehicles and

many of the vehicles are assigned to multiple plants.

Again, this is not as expensive as a full flexibility, but

it almost gives you all of the benefits.

As another example of this idea of of partial flexibility,

consider the case of software developers.

As a manager hiring a pool of software developers,

ideally, you would love to have software developers that are able to do everything.

Program in C++, SQL, graphics designs, Drupal, whatever it might be.

However, such fully flexible developers are rare and would be very expensive.

The idea of partial flexibility is that you would hire people who master two

areas of work, say, SQL and C++.

You want to make sure that everybody you hire would at least

be able to do two different things for you, and you make sure that for every

domain where you need software development expertise, you hire at least two people.

This gives you a partial flexibility at dramatically lower cost.

Instead of optimizing the batch sizes, given the set-up time,

why not challenge the set-up times themself?

After all, it is the set-up times that are the root cause for

all the problems that we've encountered so far in this module.

In this session, we talked about the concept of SMED.

SMED stands for Single-Minute Exchange of Dies, and

it is a very powerful way to cut set-up times.

You can apply SMED to production settings but you can also apply SMED to reducing

the changeover time in an operating room or the gate time for an aircraft.

We also saw how you can configure a network of plants in a way that takes

advantage of pooling the amount uncertainty.

We saw that a little bit of flexibility often goes a very long way,

especially if you change the plants, and

then reconfigure the plant network in a clever way.

Explora nuestro catálogo

Inscríbete de manera gratuita y obtén recomendaciones personalizadas, actualizaciones y ofertas.

Coursera brinda acceso universal a la mejor educación del mundo, al asociarse con las mejores universidades y organizaciones, para ofrecer cursos en línea.

© 2018 Coursera Inc. Todos los derechos reservados.

Descargar en la App Store

Consíguelo en Google Play