Professional Documents
Culture Documents
Solutions To Sample Exam Questions
Solutions To Sample Exam Questions
Question 1
The city of London has just decided to build a new underground station on the Jubilee
Line, between Baker Street and St Johns Wood. The new station located in St
Johns Wood Road, just opposite the Lords Cricket Ground, will only serve the
Jubilee line.
Matt Richardson is in charge of designing the layout for this new station and he is
currently dealing with the placement of escalators. Due to space constraints the station
can accommodate only three escalators. Hence, two configurations are possible, as
indicated in the figure below.
railway
railway
platform
platform
hall
hall
street
street
CONFIGURATION A:
CONFIGURATION B:
1 UPWARD EXCALATOR
2 UPWARD EXCALATORS
2 DOWNWARD EXCALATORS
1 DOWNWARD EXCALATOR
Configuration B
Why?
The problem can be analyzed by considering the tube station as a combination of two
queuing systems, where the escalators are the servers. In the first system incoming
customers queue to access the downward escalator, whereas in the second system
outgoing customers queue to reach the hall upstairs and exit the station.
The objective of the designer should clearly be to maintain the average queue length
below an acceptable level. Given that the platform and the hall have the same
surface they can tolerate approximately the same number of customers waiting, hence
the acceptable queue length is the same in the two cases. Matt should therefore
analyze the two systems (upward and downward) when operated with only one
escalator, and allocate the additional escalator to the one that is more likely to
generate the longest queue.
Remember that queue length is proportional to scale, utilization and variability.
Clearly in this case scale and utilization are the same for the two systems (same
escalator speed and same amount of incoming and outgoing customers). Conversely,
the variability of outgoing customers is larger than the variability of incoming
customers. To see this consider that whereas incoming customers arrive at the hall
one by one, following a poisson process (coefficient of variation = 1), outgoing
customers arrive at the platform in large batches, whenever a train stops. This is a
bursty arrival process and has a coefficient of variation larger than one.
Therefore, everything else being equal the system in which customer queue to exit the
station is likely to generate the longest queue and should be given the additional
escalator.
Question 2
Newsboy losses are a serious concern for many firms in the fashion industry. Which
one between the two companies below is more likely to suffer from these losses?
Why?
X
Zara
Answer:
Newsboy losses are a serious concern for companies that because of their long lead
time - need to forecast demand ahead of time. This is the case for Marks & Spencer,
which have on average much longer lead times than Zara. Alternatively one could
argue that Marks & Spencer operate in a much more stable market than Zara
(traditional UK customers as opposed to young urban shoppers). Hence in theory
it would be easier for them to predict demand and to hedge against newsboy losses.
This argument is less strong than the first one, though.
Question 3
You are part of the operations management team at Eric & Old, a major consulting
company.
As a first assignment your group is brought in to help Rodi improve their customer
response times to their new customers in the automobile industry. Rodi manufactures
components for airplanes and motor vehicles. Until now, they have mostly worked in
the defense sector, where customers are more concerned with technology and quality
than with response time. In the new environment response time and production cost
are much more important and have become a key priority to them.
Why?
Clearly, the system is constrained by the upstream workstation, which is the real
bottleneck, not by the downstream machine. Changing the schedule on the upstream
machine will not have an impact on the utilization of the downstream machine,
because on average the latter will always see the arrival of the same number of jobs
(determined by the throughput of the bottleneck workstation).
Therefore, the only way to increase the utilization rate of the downstream machine is
to increase the throughput at the bottleneck. However, the larger number of set ups
required by the new policy is likely to have to opposite effect: it will waste valueadded time at the bottleneck and further constrain the productivity of the whole
factory.
As a result, at best the policy will have no effect. Most likely it will even make the
situation worse.
As a general remark, you should suggest to your client that maximizing utilization at
a non-bottleneck machine is a wrong objective. You should advise Rodi to manage the
system to achieve maximal throughput, not maximal utilization of machines.
Question 4
In a maketo-order environment, such as the one you encountered in the Littlefield
simulation - the only viable strategy to reduce lead-time is to increase capacity by
adding more machines to each station.
Yes
NO
WHY?
Reducing lead time is equivalent to reducing waiting time in a queue. This objective
can be achieved also by reducing variability in the system and by increasing the pace
of operations (for instance replacing slow machines with faster workstations) to
reduce utilization.
Question 5
a) 10 points
You have been appointed as the new quality director of Forevon, a California-based
manufacturer of CCD sensors for digital cameras. Forevons customers which
include some of the largest camera manufacturers in the world understandably
require high quality standards from their sensor supplier. One of the most important
parameters is the sensor size, which must fit the design specifications provided by the
camera manufacturers. To guarantee the high quality standards that their customers
expect, Forevon has implemented a quality control system that makes extensive use of
SPC techniques such as control charts. You have just received a control chart with the
results of the last conformance analysis conducted on the production process of one of
Forevons flagship products, the 4 megapixel sensor. The analysis suggests that the
process is stable and in control: the mean and the standard deviation of both the
sensor length and height have not changed significantly over the last 3 weeks. Can
you rely on this information to conclude that no improvements to the process are
needed and that the process is capable of meeting the customer requirements?
yes
no
Why?
The information provided by the control charts only suggests that the process is in
control (i.e. its underlying distribution is not shifting). However it doesnt say
anything about its capability. In other words the process could still generate an
unacceptable number of defective items (although it would do so quite consistently...).
To conclude that no immediate improvement action is needed, the manager should
also compute a capability index and conduct a capability analysis to check whether
the process delivers outputs in accordance to customer requirements.
b) 10 points
Forevon (see question above) has recently launched a new product, a 6 megapixel
sensor with an above-average color rendition, offered at a particularly competitive
price. Due to the high market demand, the sensor is manufactured in two parallel and
identical production lines, A and B. As a quality director you have just received the
control chart for the two processes. The charts suggest that both lines went out of
control. Although it maintained its usual level of variability, the production process on
Line A had a significant shift in its mean, which caused a large number of sensors to
fall outside the specified customer tolerance. Conversely, the production process
associated with Line B didnt display any significant shift in its mean. However it did
have an increase in variability, which also caused an equally large number of products
to fall outside the specified customer tolerance. You must obviously react to this
situation. Which of the two lines is likely to be easier to bring back in control?
X
Line A
Line B
Why?
The problems occurred in line B are more difficult to identify because they seem to be
due to random or natural variation. In other words there are no clear signals that
can help the manager identify the source of the problem. Most likely the deterioration
of quality is caused by the simultaneous degradation of a number of factors.
Conversely, the problems occurred to line A are clearly due to an assignable cause
which determines a systematic bias in the process. Assignable causes are much
easier to identify and should immediately trigger the implementation of a PDCA
cycle.
Question 6
Draw a product-process matrix for the apparel industry and use it to discuss the
operations strategy of the following companies: i) Marks & Spencer; ii) the haute
couture department of a high fashion company; iii) a mass merchant; iv) Zara
Solution:
Product
High
customisation
Process
Some
customisation
High
standardisation
High fashion
Manual
shop
Batch
flow
Rigid Line
flow
(Dior, Chanel, )
ZARA
M&S
Mass
merchants
The haute couture department of a high fashion company produces highly customized
products in extremely low volumes. In order to achieve such a high level of
customization the production process must be structured as an extremely flexible (and
costly) job shop.
Mass merchants rely on process standardization and economies of scale to produce
large volumes of low-cost, highly standardized products.
Question 7
Do you agree with the following statement?
The increased inventory management costs caused by inexact demand forecasts are
more severe for a push production system than for a pull production system.
X
Yes
No
WHY?:
In a push production system, the production schedule of each station is set ahead of
time and depends on the estimated demand. A push system adjusts to demand
fluctuations by either consuming or building up inventory. Conversely, a pull
production process is a self-regulated system in which production is authorized when
the level of inventory before each station (i.e. Work-In-Progress) falls below a
predetermined threshold. A pull environment reacts to fluctuations in demand by
automatically slowing down or speeding up operations while maintaining a constant
level of WIP in the system. Therefore, a push system in more likely than a push
system to incur in higher inventory costs when demand forecasts are inaccurate.
Question 8
1. Sport Obermeyer sources a particular parka from a supplier in China at a (total
landed) cost of $60. The delivery lead time is 8 months, and therefore Sport
Obermeyer needs to source it well before the selling season. The demand is for
this item over the selling season is normally distributed with mean 2000 and
standard deviation 600. How many units should Sport Obermeyer buy, if it can
sell the item for $120 during the regular selling season and at a 60% discount at
the end of the selling season (assuming Sport Obermeyer wants to maximize
expected profits)?
Answer:
The order quantity Q* can be determined from the usual formula: Q* = + z
where z is the the z-score corresponding to the optimal service level:
p* = Cu/(Cu+Co)
Question 9
For the process shown below :
M3
12 min
M1
3 min
M2
5 min
M4
10 min
M5
15 min
M6
4 min
[To achieve the minimum TPT, the transaction will go through the stages using the fastest
machine available at each stage. So its M1M2M4M6]
[When computing the cycle time at stage consisting of parallel machines, one needs to
convert the CT (cycle time) first into capacity and then add up the capacities of each
individual machine to have the total capacity. You shall then convert this total capacity back
to CT. Remember for parallel machines, only capacity can add up but not the CTs.]
Question 10
You have been hired as operations consultant by a manufacturing firm that assembles
LCD monitors. The company has recently launched a new flagship product, the
20monitor XPS-20. Assembly takes place onto a dedicated line composed of exactly
4 workstations operating in series, with no intermediate buffer between them. The
first 3 workstations in the line (1, 2 and 3) have exactly the same capacity (say 10
units/hour on average). Station 4 has a slightly lower capacity (say 8 units/hour, on
average). Despite the efforts made by the management to stabilize flow, all stations
have some output variability (i.e. the actual output of each station can be above or
below its average capacity).
Figure 1
The company is concerned that the actual productivity of the line does not seem to be
sufficient to meet the demand for the XPS-20 monitor. To increase the throughput of
the process, the management is considering adding some WIP buffers to prevent
downstream machines from starving. However, due to space limitations and budget
constraints, the company can afford to add only one buffer of a given size to the
whole line. The buffer can be added between either stations 1 and 2, 2 and 3 or 3 and
4 (see figure 1). The operations management team spent long hours discussing the
pros and cons of each of the three alternatives (1-2 vs 2-3 vs 3-4), but could not reach
a conclusion. Mindful of the OM knowledge that you have recently acquired during
your OM course at HEC Paris, which option would recommend?
a) Place buffer between station 1 and station 2
b) Place buffer between station 2 and station 3
c) Place buffer between station 3 and station 4
Answer: c)
As the capacity of the process is determined by the bottleneck capacity the buffer
should be positioned in front of the bottleneck. Without a buffer, the workstations are
not independent. That is, when the output of the upstream machine is lower than the
bottleneck theoretical output, the bottleneck would starve. The WIP buffer positioned
in front of the bottleneck would work as decoupling stock and make the bottleneck
independent from the upstream workstations.
Question 11
The Toyota Production System (TPS) is widely recognized as one of the most
effective and most efficient approaches to manufacturing management, especially in
the automobile industry. List and discuss at least three major characteristics of this
system.
Answer:
Some of the most important characteristics of TPS are:
Holistic approach: optimization of the system, not of the individual parts
Focus on process control: use of standard and simple procedures to reduce
variation
Continuous improvement: problems are natural sources of learning, use of the 5whys approach;
Jidoka: highlight problems when they occur
Elimination of waste
Heijunka: value chain from customer to suppliers, involvement of suppliers
Use of JIT (Kanban) as a flow management discipline - limited (zero) inventory
Workers involvement and empowerment: Andon cords to stop the line, training
and coaching
Question 12
Everything else being equal, in a continuous review inventory model the optimal
order quantity q increases when... (tick all answers you deem correct)
a) The inventory holding cost decreases
b) The ordering cost decreases
c) The annual demand decreases
d) None of the above
Answer:
Clearly a), from the EOQ formula: Q* = SQRT (2SD/H)
Question 13
Specialty Coffeehouse
A specialty coffeehouse sells Colombian coffee at a fairly steady rate of 280 pounds
annually. The beans are purchased from a local supplier for $2.40 per pound. The
coffee house estimates that it costs $45 in paperwork and labor to place an order for
the coffee, and holding costs are based on a 20 percent annual interest rate.
a. Determine the optimal order quantity for Colombian coffee.
b. What is the time between placement of orders?
c. What is the average annual cost of holding and setup due to this item?
d. Determine the reorder level (on-hand inventory at time of reorder) if the
replenishment lead-time is three weeks
Answer:
(a) EOQ = 229.13 pounds
(b) Time between orders = 0.82 years
(c) Average annual setup and holding cost = $109.98
(d) Reorder level = 16.15 pounds