MAN 305 Operations Research II

Employee Scheduling Problem

Air-Express is an express shipping service that guarantees overnight delivery of packages anywhere in
the continental United States. The company has various operations centers, called hubs, at airports in
major cities across the country. Packages are received at hubs from other locations and then shipped
to intermediate hubs or to their final destinations.

The manager of the Air-Express hub in Baltimore, Maryland, is concerned about labor costs at the hub
and is interested in determining the most effective way to schedule workers. The hub operates seven
days a week, and the number of packages it handles each day varies from one day to the next.

Using historical data on the average number of packages received each day, the manager estimates
the number of workers needed to handle the packages as given in the table.

Day of the week Workers Required

Sunday 18
Monday 27
Tuesday 22
Wednesday 26
Thursday 25
Friday 21
Saturday 19

The package handlers working for Air-Express are unionized and are guaranteed a five-day work week
with two consecutive days off. The base wage for the handlers is $655 per week. Because most
workers prefer to have Saturday or Sunday off, the union has negotiated bonuses of $25 per day for
its members who work on these days. The possible shifts and salaries for package handlers are given
in the table.

Shift Days off Wage

1 Sunday and Monday $680
2 Monday and Tuesday $705
3 Tuesday and Wednesday $705
4 Wednesday and Thursday $705
5 Thursday and Friday $705
6 Friday and Saturday $680
7 Saturday and Sunday $655

The manager wants to keep the total wage expense for the hub as low as possible. With this in mind,
how many package handlers should be assigned to each shift if the manager wants to have a sufficient
number of workers available each day?

Modelling of the Problem

One of the problem areas of linear programming is employee planning. In these problems, we aim to
assign the employees to the shifts or days in order to keep the employee cost as low as possible. The
constraints are written to ensure that there are enough number of employees in each shift to keep
the operation running.

In Air-Express employee scheduling problem, we have a hub that works 7 days of the week. Every day
there are certain number of employees to be on duty to handle the packages given in the first table.

1
MAN 305 Operations Research II

The employees work in shifts and shifts consist of 5 days of work time. This means that every shift has
2 days off. These days are consecutive as given in the second table. An employee assigned to shift 1
works Tuesday to Saturday and takes Sunday and Monday off. An employee assigned to shift 2 works
Wednesday to Sunday for five days and so on. We have the cost per employee for each shift in the
second table.

In this problem, we want to answer the following question: “How many employees should be assigned
to each shift so that we have enough employees every day and our total personnel cost is minimum?”

Decision Variables

We have 7 shifts to assign the employees and therefore we have 7 decision to make. We can write the
decision variables as below

𝑥" : the number of employees assigned to shift 1

𝑥# : the number of employees assigned to shift 2
𝑥$ : the number of employees assigned to shift 3
𝑥% : the number of employees assigned to shift 4
𝑥& : the number of employees assigned to shift 5
𝑥' : the number of employees assigned to shift 6
𝑥( : the number of employees assigned to shift 7

We could also write the decision variables in the closed form as below.

𝑥) : the number of employees assigned to shift 𝑖 𝑖 = 1,2, … ,7

Objective function

The company wants to minimize its employee cost. We have to write down a total cost function by
using the wages paid to each worker in the shifts and the number of workers in each shift that is
defined with 𝑥 variables.

𝑀𝑖𝑛 680𝑥" + 705𝑥# + 705𝑥$ + 705𝑥% + 705𝑥& + 680𝑥' + 655𝑥(

Note: Let us also explain why wages change between shifts. As given in the problem, “the base wage
for the handlers is $655 per week. Because most workers prefer to have Saturday or Sunday off, the
union has negotiated bonuses of $25 per day for its members who work on these days”. Depending on
the shift including any weekend days or not the wage is changing. The amounts in the second table
are the calculated amounts. For example, because Shift 7 does not include any weekend day, the wage
per person in this shift is $655 per person. On the other hand, Shift 1 includes one weekend day
(Saturday). Therefore, the wage is $655+$25=$680 per person. Shift 2 includes both Saturday and
Sunday as work days. Therefore, the wage is $655+$50=$705. This is the reason why the wages change
between shifts.

2
MAN 305 Operations Research II

Constraints

Þ Employee Requirement Constraints. In employee scheduling problems with different work days
depending on the shift, it is very useful to make a table of “who works when?”. Such a table will
help us to write down the constraints to satisfy the requirements each day. The hub works for 7
days per week and every day we need certain amount of employees to be at work. For example,
on Sunday there should be at least 18 workers at work to keep the operation running. In order to
write down a constraint for Sunday, we have to know the number of working employees on
Sunday in terms of 𝑥 variables. The below table will assist us.

Number of
Shift Sun Mon Tue Wed Thu Fri Sat
Workers
1 0 0 1 1 1 1 1 𝑥"
2 1 0 0 1 1 1 1 𝑥#
3 1 1 0 0 1 1 1 𝑥$
4 1 1 1 0 0 1 1 𝑥%
5 1 1 1 1 0 0 1 𝑥&
6 1 1 1 1 1 0 0 𝑥'
7 0 1 1 1 1 1 0 𝑥(

The table has shifts on the rows and the days of week as columns. We write 1 in the corresponding
cells if the employees in shift 𝑖 work in shift 𝑗 and 0 if not. For example, Shift 1 workers take Sunday
and Monday off. Therefore we have 0 values for Sunday and Monday in Shift 1 row. They work
Tuesday to Saturday, which is represented by 1 in the Shift 1 row. Workers assigned to Shift 2 take
Monday and Tuesday off. It means that they work from Wednesday to Sunday.

The last column of the table presents the numbers of workers in the shifts in terms of 𝑥 values. To
write down the constraints for every day (we will have 7 constraints for 7 days), we can read the
table vertically. When we look at the Sunday column, we can see workers of which shifts are on
duty on Sunday by observing the values of 1.

On Sunday, workers of Shifts 2,3,4,5 and 6 are working. The workers of Shifts 1 and 7 are absent
on this day. Therefore, we can calculate the number of employees that are at work by multiplying
the coefficients in Sunday column with the values in the column for number of workers in each
shift. Below is the number of employees working on Sunday:

0𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 0𝑥(

This total should be at least 18 since 18 workers are required on Sunday. So, we have:

Sunday Constraint: 0𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 0𝑥( ≥ 18

On Monday, the workers of Shifts 3,4,5,6 and 7 are working. The workers of Shifts 1 and 2 are
absent on this day. Therefore, we can calculate the number of employees that are at work by
multiplying the coefficients in Monday column with the values in the column for number of
workers in each shift. Below is the number of employees working on Monday:

0𝑥" + 0𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 1𝑥(

This total should be at least 27 since 27 workers are required on Monday. So, we have:

3
MAN 305 Operations Research II

Monday Constraint: 0𝑥" + 0𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 27

With the same logic, we can write down the constraints for the rest of the days.

Tuesday Constraint: 1𝑥" + 0𝑥# + 0𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 22

Wednesday Constraint: 1𝑥" + 1𝑥# + 0𝑥$ + 0𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 26
Thursday Constraint: 1𝑥" + 1𝑥# + 1𝑥$ + 0𝑥% + 0𝑥& + 1𝑥' + 1𝑥( ≥ 25
Friday Constraint: 1𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 0𝑥& + 0𝑥' + 1𝑥( ≥ 21
Saturday Constraint: 1𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 0𝑥' + 0𝑥( ≥ 19

Þ Non-negativity Conditions. Finally, we have non-negativity constraints that will ensure all 𝑥values
to be either zero or positive, since we cannot assign negative people in shifts.

𝑥" , 𝑥# , 𝑥$ , 𝑥% , 𝑥& , 𝑥' , 𝑥( ≥ 0

Þ Integrality Conditions. The decision variables should take integer values because they represent
the number people.

𝑥" , 𝑥# , 𝑥$ , 𝑥% , 𝑥& , 𝑥' , 𝑥( 𝑖𝑛𝑡𝑒𝑔𝑒𝑟

Full Model

When all of above come together, we have the below LP model to solve for obtaining the optimal
number of employees to be assigned in the shifts with minimum cost.

𝑀𝑖𝑛 680𝑥" + 705𝑥# + 705𝑥$ + 705𝑥% + 705𝑥& + 680𝑥' + 655𝑥(

𝑠𝑢𝑏𝑗𝑒𝑐𝑡 𝑡𝑜
0𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 0𝑥( ≥ 18
0𝑥" + 0𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 27
1𝑥" + 0𝑥# + 0𝑥$ + 1𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 22
1𝑥" + 1𝑥# + 0𝑥$ + 0𝑥% + 1𝑥& + 1𝑥' + 1𝑥( ≥ 26
1𝑥" + 1𝑥# + 1𝑥$ + 0𝑥% + 0𝑥& + 1𝑥' + 1𝑥( ≥ 25
1𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 0𝑥& + 0𝑥' + 1𝑥( ≥ 21
1𝑥" + 1𝑥# + 1𝑥$ + 1𝑥% + 1𝑥& + 0𝑥' + 0𝑥( ≥ 19
𝑥" , 𝑥# , 𝑥$ , 𝑥% , 𝑥& , 𝑥' , 𝑥( ≥ 0
𝑥" , 𝑥# , 𝑥$ , 𝑥% , 𝑥& , 𝑥' , 𝑥( 𝑖𝑛𝑡𝑒𝑔𝑒𝑟