Caso de Estudo 3

Assigning Students to Schools*

Investigacao Operacional I
Mestrado Integrado em Engenharia Industrial e Gestao
Ano Lectivo 2007/2008
14 de Outubro de 2008

* O caso de estudo Assigning Students to Schools foi adaptado do caso original

apresentado no livro: Hillier, F.S. and Hillier M.S., Introduction to Management
Science: A Modelling and Case Studies Approach with Spreadsheets, McGraw-Hill, 2003.

Assigning Students to Schools

The Springfield School Board has made the decision to close one of its middle schools
(sixth, seventh, and eight grades) at the end of this school year and reassign all of next
years middle school students to the three remaining middle schools. The school district
provides bussing for all middle school students who must travel more than approximately
1 km, so the school board wants a plan for reassigning the students that will minimize
the total bussing cost. The annual cost per student for bussing from each of the six
residential areas of the city to each of the schools is shown in the following table (along
with other basic data for next year), where 0 indicates that bussing is not needed and a
dash indicates an infeasible assignment.
Bussing cost per student
Area

No. of students

% in 6th grade

% in 7th grade

% in 8th grade

School 1

School 2

School 3

450

32

38

30

600

37

28

35

e300

e700

e400

e500

550

30

32

350

28

40

38

e600

e300

e200

32

e200

e500

500

39

34

27

450

34

28

38

e400

e500

e300

900

1.100

1.000

School capacity:

The school board also has imposed the restriction that each grade must constitute
between 30 and 36 percent of each schools population. The above table shows the
percentage of each areas middle school population for the next year that falls into each
of the three grades. The school attendance zone boundaries can be drawn so as to split
any given area among more than one school, but assume that the percentages shown in
the table will continue to hold for any partial assignment of an area to a school.
You have been hired as a management science consultant to assist the school board
in determining how many students in each area should be assigned to each school.
1. Formulate a linear programming model for this problem.
2. Solve the model. What is your resulting recommendation to the school board?

3. Before adopting a busing plan based on part (2), the school board now wants to
conduct some sensitivity analysis.
(a) Use the solver to generate the sensitivity report for the model solved on part
(2).
(b) One concern of the school board is the ongoing road construction in area 6.
These constructions projects have been delaying traffic considerably and are
likely to affect the cost of busing students from area 6, perhaps increasing costs
as much as 10%.
Use the sensitivity report to check how much the busing cost from area 6 to
school 1 can increase (assume no change in costs for the other schools) before
the current optimal solution might no longer be optimal. If the allowable
increase is less than 10%, use the Solver to find the new optimal solution with
a 10% increase.
(c) Repeat part (b) for school 2 (assuming no change in costs for the other schools).
(d) The school board has the option of adding portable classrooms to increase
capacity of one or more of the middle schools for a few years. However, this
is a costly move that the board would only consider if it would significantly
decrease busing costs. Each portable classroom holds 20 students and has
a leasing cost of e2.500 per year. To analyse this option, the school board
decides to assume that the road construction in area 6 will wind down without
significantly increasing the busing costs from that area.
i. For each school, use the corresponding shadow price from the sensitivity
analysis report to determine whether it would be worthwhile to add any
portable classrooms.
ii. For each school where it is worthwhile to add any portable classrooms, use
the sensitivity report to determine how many could be added before the
shadow price would no longer be valid (assuming this is the only school
receiving portable classrooms).

iii. Use the shadow price information to determine the best alternative in
terms of minimising the total cost of busing students and leasing portable
classrooms (assume that only one school can receive portable classrooms).
Use the solver for finding the corresponding optimal solution for assigning
students to schools.
4. The school board is considering eliminating some bussing to reduce costs. Option
1 is to only eliminate bussing for students traveling 1 to 1.5 km, where the cost per
student is given in the table as e200. Option 2 is to eliminate bussing for students
traveling up to 2 km, where the estimated cost per student traveling between 1.5
to 2 km is e300.
(a) Revise the model from question (1) to fit Option 1, and solve. Compare these
results with those from question (2), including the reduction in total bussing
cost.
(b) Repeat the analysis in part (a) for Option 2.
5. The school board now needs to choose among the three alternative bussing plans
(the current one or Option 1 or Option 2). One important factor is bussing costs.
However, the school board also wants to place equal weight on a second factor: the
inconvenience and safety problems caused by forcing students to travel by foot or
bicycle a substantial distance (more than 1 km, and especially more than 1.5 km).
Therefore, they want to choose a plan that provides the best trade-off between these
two factors.
(a) Use your results from questions (2), (4a) and (4b) to summarize the key information related to these two factors that the school board needs to make this
decision.
(b) Which decision do you think should be made? Why?