Name:
Program:

PROD 2100

Production and Operations Management

Exam: June 2001

Grading :

Question / 1 / 2 / 3 / 4 / 5 / 6 / 7 / Total
Grade
Over / 3 / 4 / 5 / 2 / 2 / 2 / 2 / 20

Time:1.30 – 4.30 pm

Please:

-write your name at the top of this page;

-do not unstaple this questionnaire ;

-do freely use an English dictionary;

-read each question carefully before answering it;

-use the back of the page if you need more space;

-motivate each answer. In case of ambiguity or of lack of information, first state your interpretation/assumption and then answer the question.

Have a good exam !

Question 1 ( 3 points)

You are the manager of the company MarketResearch. This company is specialized in designing and conducting market researches for consumer goods like shampoos and toothbrushes. The technique you specialized in is the public opinion poll based on a written questionnaire. The process you use for a research is rather standard. It is made of the following successive operations:

  1. Understanding the product features
  2. Understanding the goal of the marketing research
  3. Designing and formulating an adequate questionnaire
  4. Pre-testing the questionnaire
  5. Administrating the questionnaire to a sample of customers by individual interviews (usually around 500 interviews are performed)
  6. Encoding the results in order to perform a statistical analysis of these results
  7. Writing the report and presenting it to the customer

You could choose between two types of organizations. Either, something like a line or a synchronized job-shop in which every step is done by a different person or; something more like an individual organization in which all the operations of a same research are done by a single person.

Question.For this particular case (and not in general):

a)what are the main differences (strengths and weaknesses) between these two types of organizations and

b)which kind of organization (the first, the second or something else) would you think is most adequate ?

ANSWER:

line or job-shop:+narrow-skill workforce (specially for 5 or for the encoding of the data) , specialization , learning,all this leading to low cost

- is it really possible to split all the tasks (like 1 and 2, for example) ? difficulty to control, poor customer contact

ideal: an intermediate solution: Assign the operations 1, 2, 3, 4 and 7 to the same person. Assign 5 to another person with the right skills. Also for 6 if it requires “stat” skills or “typing skills”.

Question 2 (4 points)

(The answer to the following question should be based on the experience you developed with the Picasso case.)

You are the person responsible for the organisation of an academic year, let us say, SESP11 or INGE11. The program approved by the university authorities is made of 12 different courses, 6 per semester.

Assume now that the university has decided that new pedagogical principles (e.g. active participation of the students, explicit use of cases, group work, use of new technologies, ...) should be followed. Being responsible for this program year, you have asked each professor in charge of a course to propose a new structure for his/her part. The result was quite positive. Each of the 12 professors came with a new course structure including lectures, exercises, articles to read, cases to be solved, oral presentations and exams.

However, your problem is now to determine :

a)whether the students will not be overloaded by the work requested by all these courses;

b)whether you have enough capacity in terms of computer facilities;

c)a feasible schedule for all the different activities requested by the new structures of the courses.

Question: Explain how you will proceed to solve this problem. Sketch precisely what you will compute and which data you will need.

ANSWER:

- for each product(course): get the amount of resource that is required (in terms of student work time and computer facilities) and the deadlines if any (otherwise, we assume that the end of the year is the deadline).

- aggregate the demand (use for example the following aggregate units: the student’s working hour and the computer hour) to know the total required capacity.

get the available capacities (in terms of student work time and computer facilities)

check whether there is enough capacity for the total aggregated demand. If there isn’t, then stop. (answer to question a and b)

If there is, try to find a production plan without backlog (it is most reasonable to base the plan on the student’s work hour). If you need some backlog, it means that some of the deadlines are too early. Some deadlines must then be postponed.

Disaggregate the production plan in order to determine a feasible schedule. There are many options here for the disaggregation. The “computer room” capacity can constraint the disaggregation. (answer to question c)

Question 3 (5 points)

You are controlling the inventory of the good A in a hospital that is operating 365 days a year.

There is only one supplier for good A. He offers regular and emergency deliveries. Regular deliveries are scheduled on Friday morning (every week if you wish). However, the regular order must be placed Wednesday morning at the latest if you want to be delivered on the next Friday. Besides these regular orders, he also offers you an “emergency order service” by which he can supply you within 2 hours. Emergency orders can be placed 24 hours a day, 7 days a week. However, this service is expensive and furthermore, he can only deliver one unit a time for capacity reasons.

You would like to place regular orders at an adequate frequency and, in case of a shortage, to place an emergency order. Indeed, if you face a shortage you cannot simply wait for the next regular deliveries and the 2 hour delay is acceptable for you.

Here are some data about this product and about the delivery system. Costs are expressed in Euro’s (E)

Cost of one unit of good A: 1000 E

Transport cost for a regular order (independent of the quantity) : 60 E

Transport cost for an emergency order : 500 E

Administration cost (paper work, reception and control of the goods) : 20 E per order

(for regular and emergency orders)

Yearly storage cost (warehouse, insurance, ...) : 20 % of the stored value

Yearly opportunity cost of the money: 20 %

Daily demand for good A:average: 1 unit

Standard deviation: 1 unit

Question:

By neglecting safety stock considerations, you would like to place regular orders in an economical way.

  1. Explain how you will proceed;
  2. If you have all the needed data, compute how frequently you would place regular orders, taking into account the features of the supply system.
  3. In the questions 4 to 6 here below, an adequate safety stock ss will be computed. Assuming here ss to be known, which inventory control system would you then suggest and with which parameters?

You would also like to keep an adequate safety stock.

  1. Explain which objective you would aim at;
  2. Explain how you will proceed to reach this objective;
  3. If you have all the needed data, compute that adequate safety stock, taking into account the features of the supply system.

The formulas here below could be of some help (but certainly not for questions 1,3,4 and 5).

Q = ;
minimize RProbability[DLt > R] 

Q = ;
minimize Rfill rate:

if backorder : = ( Q - n(R) ) / Q

if lost sales : = Q / ( Q + n(R) )

Stockout Probability = Prob[ D(Vp) > Ii + Qi ]

Assume D(Vp) is N((Vp),(Vp))

Choose: Qi = Vp + k Vp - Ii = Qmax - Ii

ANSWER:

For 1. Find the economical order quantity by minimizing the sum of the yearly order costs and yearly holding cost.

For 2. Q= sqrt(2 (60+20) (365) / (0.40 1000))= sqrt( 4 36.5) = 12 that is just less than the need for 2 weeks. I would therefore order every two weeks.

For 3. (periodic system) Every second Wednesday morning, place an order of size : (Qmax-I(observed)), with Qmax = 14 + 2 + ss

For 4. Since we use the emergency orders in case of shortages, we will aim at an economical objective, i.e. at a situation that minimizes the sum of the costs due to emergency orders and of the holding cost of the safety stock;

For 5. We will proceed by a marginal analysis. We will bring the safety stock to a level such that : the holding cost of one more unit of goods should be equal to the emergency costs saved by this one more unit.

For 6.H(1 unit) = 400 = shortage prob * 26 * 520;
=> shortage prob = 0.03
Vp = 2 weeks + 2 days = 16 days
demand in (16 days) = (16*1 unit ; 4*1 unit) assumed to be “normally distributed”.
shortage prob = 0.03 => 2 std dev = 8 units = SS

Question 4 (2 points)

Which weaknesses of the MRP approach is the Just-in-time concept trying to improve?

ANSWER:

The MRP techniques aims at living (not too badly) with several problems such as, for example:

setup or order cost->taken care by lot sizing techniques

long delay->taken care of in the MRP record

quality problems->taken care by safety stock in the MRP record

unreliable supply->taken care by safety time

However, these problems are not solved. In addition, the MRP technique itself aims at solving the question of “controlling” the organization (who is doing what) by means of a huge administration.

The JIT aims at going towards a line organization, that is an organization with: no or reduced setups and delays; no or reduced quality problems, reliable processes, no need for this paper control (kanban instead of reports), and so on.

Question 5 (2 points)

The quality of the layout of a factory or of a service facility has a great impact on the performance.

Question:

  1. Propose a clear characterization of a “good” layout and motivate it (several answers are possible)
  2. Explain how you would proceed to improve the layout of a facility (which data would you need and how would you use it)

ANSWER:

1. For a factory, a good layout would definitely be one that reduces as much as possible the handling of material. For service facility, a good layout would most often aim at reducing the distance traveled by the customers.

(other answer) In a department store, an additional goal could be aimed at: that the customer crosses the whole shop in order to expose him as much as possible to the goods on sales.

In order to reduce the handling of the material, the needed data are: what are the products, in which shop do they require service and how many do we produce. This allows us to define the need in terms of “how many trips per day from which shop to which shop” are required. Based on these data, we can try to minimize the “traveled distance” by defining where the different shops will be located. For example, if all the products that require service in shop A do also require afterwards service from shop B, then we would minimize the traveled distance by having the shop B besides the shop A.

Question 6 (2 points)

Since several years you use a stationary model for estimating your weekly needs in material A. In fact, this model is justified by the fact that you never detected any seasonality or trend for the need for this material. In order to get the parameters for your model, you used the exponential smoothing method with ( = 0.20).

You are facing the week 21 of this year. Here are all your last data:

D(20) = 20; F(20) = 30;

You also record your errors since the beginning of the year. Here are the results:

MAD = 3; TSE = 1;

Question: Based on these data, argue whether

-you should change the model

-you should change the method

-you should change its parameter 

MAD = TSE =

ANSWER:

There is no observable bias (TSE is very small).
The MAD is rather small (a 10 % error on the average).
About the model, there is no need to incorporate some “trends”. Indeed, if there is a trend and you use a stationnary model, you will always be trailing below or above the observations. This would result in a large TSE. Since it is not the case, we can assume thet there is no trends.

About the model, we do not have any information to verify that there is no seasonality.. A model that incorporates a seasonality could be tried.

The method (exponential smoothing) is ok for the model.

Alpha is still quite large for a system that is in operations since long. It could maybe be decreased.

Question 7 (2 points)

In your opinion, which management concept do you see at the intersection of

a)the paper “A conceptual Model of Service Quality and Its Implications for Future Research" by Parasuraman, Zeithaml and Berry;

b)the principles of the ISO-9000 norms, and

c)“the management observation checklist used to evaluate the performance of counter staff in fast food restaurant” discussed in the chapter “Mechanization Takes Command” in the book of G. Morgan;

Summarize also briefly how each of these three sources “sees” this concept.

ANSWER:

At the intersection of these three sources we find the concept of “task standardization”.

In (a), the authors perceive the lack of “task standardization” as a potential cause (among others) for the existence of gap 3 (difference between the specifications of a service and service that is really delivered)

In (b), one of the principles of the ISO-9000 norms is “say what you do”, that is, “describe clearly how your process works”. The basic idea is that you cannot reach quality (in fact you cannot reach timely consistency) if you do not have a clearly defined process. The ISO-9000 norms requires thus the processes to be defined and these definitions to be followed. Defining the process is thus, in some way, a kind of task standardization.

In (c), the organization is compared as a machine whose basic principle is to have clearly defined parts (counter staff) with clearly defined functions (how he should proceed).

The underlying concept behind task standardization is of course “quality”.

Distribution normale N(0,1)

z= nombre d'écarts types

P(z)= Prob [ x z ] =

E(z)= nombre moyen de manquants =

z / P(z) / E(z) / E(-z)
0,0000 / 0,5000 / 0,3989 / 0,3989
0,1000 / 0,4602 / 0,3509 / 0,4509
0,2000 / 0,4207 / 0,3069 / 0,5069
0,3000 / 0,3821 / 0,2668 / 0,5668
0,4000 / 0,3446 / 0,2304 / 0,6304
0,5000 / 0,3085 / 0,1978 / 0,6978
0,6000 / 0,2743 / 0,1687 / 0,7687
0,7000 / 0,2420 / 0,1429 / 0,8429
0,8000 / 0,2119 / 0,1202 / 0,9202
0,9000 / 0,1841 / 0,1004 / 1,0004
1,0000 / 0,1587 / 0,0833 / 1,0833
1,1000 / 0,1357 / 0,0686 / 1,1686
1,2000 / 0,1151 / 0,0561 / 1,2561
1,3000 / 0,0968 / 0,0455 / 1,3455
1,4000 / 0,0808 / 0,0367 / 1,4367
1,5000 / 0,0668 / 0,0293 / 1,5293
1,6000 / 0,0548 / 0,0232 / 1,6232
1,7000 / 0,0446 / 0,0183 / 1,7183
1,8000 / 0,0359 / 0,0143 / 1,8143
1,9000 / 0,0287 / 0,0111 / 1,9111
2,0000 / 0,0228 / 0,0085 / 2,0085
2,1000 / 0,0179 / 0,0065 / 2,1065
2,2000 / 0,0139 / 0,0049 / 2,2049
2,3000 / 0,0107 / 0,0037 / 2,3037
2,4000 / 0,0082 / 0,0027 / 2,4027
2,5000 / 0,0062 / 0,0020 / 2,5020
2,6000 / 0,0047 / 0,0015 / 2,6015
2,7000 / 0,0035 / 0,0011 / 2,7011
2,8000 / 0,0026 / 0,0008 / 2,8008
2,9000 / 0,0019 / 0,0005 / 2,9005
3,0000 / 0,0014 / 0,0004 / 3,0004
3,1000 / 0,0010 / 0,0003 / 3,1003
3,2000 / 0,0007 / 0,0002 / 3,2002
3,3000 / 0,0005 / 0,0001 / 3,3001
3,4000 / 0,0003 / 0,0001 / 3,4001
3,5000 / 0,0002 / 0,0001 / 3,5001
3,6000 / 0,0002 / 0,0000 / 3,6000
3,7000 / 0,0001 / 0,0000 / 3,7000
3,8000 / 0,0001 / 0,0000 / 3,8000
3,9000 / 0,0001 / 0,0000 / 3,9000
4,0000 / 0,0000 / 0,0000 / 4,0000