Advanced Operations Management (33:623:400)
Professor Eckstein
In-Class Case Study -- Poisson Regression
You operate an online business that ships orders six days per week. For
a particular item, you have seen the following number of orders over the last
ten weeks (note that the Monday column includes orders from the preceding
Sunday):
|
Week |
Monday |
Tuesday |
Wednesday |
Thursday |
Friday |
Saturday |
|
1 |
4 |
1 |
7 |
1 |
4 |
4 |
|
2 |
3 |
1 |
2 |
1 |
1 |
5 |
|
3 |
1 |
1 |
1 |
3 |
4 |
3 |
|
4 |
7 |
2 |
0 |
3 |
6 |
6 |
|
5 |
6 |
2 |
8 |
3 |
4 |
4 |
|
6 |
3 |
3 |
5 |
4 |
7 |
5 |
|
7 |
5 |
3 |
2 |
3 |
6 |
4 |
|
8 |
3 |
2 |
4 |
4 |
3 |
3 |
|
9 |
5 |
2 |
1 |
1 |
3 |
5 |
|
10 |
2 |
1 |
2 |
0 |
5 |
5 |
You have been asked to plan a inventory policy for this item for the next
four weeks (24 business days).
- It is currently Monday, and you have 6 units in stock
- You have room to store up to 20 units; at the beginning of any day in
which you have i units of inventory, you may order at most 20
– i units.
- Holding costs are $10 per unit per day, assessed on the days ending
inventory
- Ordering new stock costs $500 per unit, plus another $500 in shipping,
handling, and overhead
- Orders arrive the next business day
- If demand exceeds your inventory, you have stock shipped directly from the
manufacturer, but that costs $650 per unit
- Value any inventory left over at the end of the four weeks at $450 per
unit.
Find the inventory policy with the least expected cost over the next four
weeks.