Homework 2, Due October 20th, 2020, Solution

Exercise 1.3

The company that is currently planning its first company owned warehouse also wants to set up a
forward zone for the 5 most commonly ordered smaller items. The head of the logistics department
estimates the parameters as follows: average number of orders per day 𝑜 = 2000, orders are picked
in batches of 10 (𝑑 = 10), travelling speed of the pickers 𝑣 = 6,000 meters per day, costs of a picker
ℎ = 100 €/day, and area and equipment costs of 𝑘 = 1 per unit of lane length have to be paid. The
other required values are provided in the table below. Calculate the number of aisle positions assigned
to each of the products, the total length of the aisles (𝑤 𝑡𝑜𝑡 ) and an estimate for the required number
of pickers (𝑁). Provide all intermediate calculations and thoughts that are needed to understand the
solution process.
𝒋 𝒐𝒋 𝒖𝒋 𝒇𝒋 𝒈𝒋 𝒘𝒋 𝒎𝒋 𝒔𝒎𝒊𝒏
𝒋 𝒔𝒎𝒂𝒙
𝒋

1 485 5 5 0,1 1 20 10 20
2 426 4 5 0,1 0,25 15 10 20
3 274 2 5 0,1 0,75 10 10 20
4 533 7 5 0,1 0,5 15 10 20
5 282 3 5 0,1 0,75 10 10 20

For the solution we need to insert the parameter values for every product into the formula:

𝑓𝑗 𝑢𝑗 𝑜𝑗
𝑠𝑗′ =
√ ℎ𝑜𝑤𝑗
𝑚𝑗 ( + 𝑘𝑤𝑗 )
𝑑𝑣

This gives us the optimal number of storage positions, but is, as you can guess, not feasible as
it is not integer. We do not know whether rounding to the next highest or to the next lowest
integer value lead to lower total costs, therefore we have to do both, and choose the integer
value which gives the lower costs (i.e. 𝑠𝑗 = ⌊𝑠′𝑗 ⌋ if 𝑐𝑗 (⌊𝑠′𝑗 ⌋) < 𝑐𝑗 (⌈𝑠′𝑗 ⌉), otherwise 𝑠𝑗 = ⌈𝑠′𝑗 ⌉). The
𝑜
ℎ 𝑤𝑗 𝑠𝑗
𝑑
costs for the storage positions for product j are calculated as 𝑐𝑗 (𝑠𝑗 ) = +
𝑣
𝑓𝑗 𝑢𝑗 𝑜𝑗
+𝑔𝑗 𝑢𝑗 𝑜𝑗 + 𝑘𝑤𝑗 𝑠𝑗 . Then, we must check if the obtained value is actually feasible, i.e. if
𝑚𝑗 𝑠𝑗
𝑠𝑗 < 𝑠𝑚𝑖𝑛
𝑗 , then the optimal number of storage positions for product j is 𝑠𝑗∗ =𝑠𝑗𝑚𝑖𝑛 , if 𝑠𝑗 >
𝑠𝑚𝑎𝑥
𝑗 , then 𝑠𝑗∗ =𝑠𝑗𝑚𝑎𝑥 , otherwise 𝑠𝑗∗ = 𝑠𝑗 . Thus, we obtain
Finally, we need to calculate the total length of the forward zone and obtain
𝑛

𝑤 𝑡𝑜𝑡
= ∑ 𝑤𝑗 𝑠𝑗∗ = 44,25
𝑗=1

The number of required workers is

𝑜 𝑡𝑜𝑡
𝑤
𝑁=⌈ 𝑑 ⌉ = ⌈1.475⌉ = 2
𝑣

Exercise 1.4

The small Austrian IT company that is currently planning its first company owned warehouse decided
to build an additional storage area for pallet-mounted server racks that need to be shipped separately.
This area will have 12 storage locations and a single I/O port. The number of required locations and
the average number of handling operations expected for each of the 3 different rack types that need
to be stored in this area is provided in the table below. The second table provides the distance from
each storage location to the I/O port. Calculate an optimal assignment of the product to storage
locations. Provide all intermediate calculations and thoughts that are needed to understand the
solution process.

𝒋 𝒎𝒋 𝒑𝒋
1 2 5
2 4 12
3 5 12

𝒌 𝒕𝒌 𝒌 𝒕𝒌
1 9 7 6
2 14 8 2
3 6 9 10
4 1 10 8
5 18 11 7
6 10 12 4
Solution: We know the number of required locations for rack type j (𝑚𝑗 ) and we know the average
number of handling operations for each rack type (𝑝𝑗 ). What we need to calculate next is how often
does each location have to be accessed. This can be simply done by dividing the average number of
handling operations of a certain rack type by the number of required locations by the same rack type.
We obtain

𝒋 𝒎𝒋 𝒑𝒋 𝒑𝒋 /𝒎𝒋
1 2 5 2,5
2 4 12 3
3 5 12 2,4
What we need to do next is to sort the rack types by their access frequency and create a table with
one entry for every location it requires, i.e. we need two entries for rack type 1, 4 for rack type 2 and
5 for rack type 3.

𝒊 (Index) 𝜶(𝒊) (access 𝜶𝒋 (𝒊) (rack

frequency for rack type assigned
type assigned to i) to i)
1 3 2
2 3 2
3 3 2
4 3 2
5 2,5 1
6 2,5 1
7 2,4 3
8 2,4 3
9 2,4 3
10 2,4 3
11 2,4 3

Then we proceed by sorting the storage location according to their length and add them to the table
above in ascending order.

𝒊 (Index) 𝜶(𝒊) (access 𝜶𝒋 (𝒊) (rack 𝜷(𝒊) (distance 𝜷𝒌 (𝒊) (storage

frequency for rack type assigned of location location number
type assigned to i) to i) assigned to i) assigned to i)
1 3 2 1 4
2 3 2 2 8
3 3 2 4 12
4 3 2 6 3
5 2,5 1 6 7
6 2,5 1 7 11
7 2,4 3 8 10
8 2,4 3 9 1
9 2,4 3 10 6
10 2,4 3 10 9
11 2,4 3 14 2

Last but not least we match storage location with rack assigned to the same index i and obtain

k k
(Location) j (Product) (Location) j (Product)
1 3 7 1
2 3 8 2
3 2 9 3
4 2 10 3
5 - 11 1
6 3 12 2