Raymond: Design Warehouse Operations - Team F

By: Dalveer Gangar, Mohak Jain, Jamie Muldoon,

Suela Peca and Madany Toure

 Critical Issue

Raymond’s distribution center is struggling to optimize their storage capacity within the

existing rack structure.

Critical Facts

Raymond’s distribution center started to notice that the customers were cancelling their

orders due to the fact that there were delays in order fulfillment. The productivity of picking and

inbound processes was declining, and the manpower requirement was at an all-time high with

overtime of approximately 5,000-man days per year. Sales were projected to grow by 40% in the

next one year which made it necessary for an expansion of storage space. However, the company

did not have enough funds to acquire land or build new racks at the Chhindwara plant.

All the zones at the distribution center together made up to 10,504 bins of which 537 bins

could only be stocked on the ground floor, specifically in zone C. Whereas, the remaining 9,967

bins were used for bolt pieces. The markets were divided into products and customer mix. The

problem was that the market scenarios were not differentiated in the software for the warehouse

management system (WMS). Which made the correct positioning of an SKU, completely

depended on the skills and experience of the workman. This also made it very challenging to

bring a new person into the team. The SKU’s were placed in bins, in each storage rack within

floor, the bins were stacked vertically over each other in levels. The exact bin piecewise capacity

was hard to distinguish due to the difference in length and fiber quality. The distribution center

didn’t have a system to define the actual number of bins that could be ascribed to a bin. So what

the workers did, they would keep scanning SKUs to a bin, even if the bin was physically full.
Since there wasn’t an actual system guided storage plan, and the whole process was dependent

on the worker knowledge and experience, if a bin selected by the worker was stuffed, he/she

would put the material on the floor and exclude the scanning procedure. Alternatively, the

worker would scan any random bin for the SKU, but without putting the actual material in it.

Which created the misplacing of the SKU’s and made it harder for the workers to find them

when they were needed to be picked up, this brought the long delays that the customers were

complaining about. Also, this process created a mess in the warehouse, because it blocked the

walkways, and led to overutilization of some bins and underutilization of some other ones, even

though the warehouse had enough capacity.

The pieces Raymond was working with, had different shapes. For instance, rolled pieces

were cylindrical whereas bolt pieces were cuboidal. At the folding area, the pieces were scanned

by using a handheld device and then they are placed in the SAP system, which meant that they

are received from folding but they were not yet stored in the warehouse bin location, the next

step would be to transfer them to the warehouse in a mixed trolley. At the warehouse the store in

a bin is identified by the put-away person, where the bin location loosely defined by the market.

The next step was optional which is to scan the piece again using a handheld device.

When it came to picking process, it took place in the warehouse in two shifts. The first

shift was from 7:30 a.m. to 3:30 p.m., and the second shift was from 3:30 p.m. to 11:30 p.m. On

an average, roughly 2,917 pieces of 12 meters each were picked on a given day. Picking

inconsistencies were made due to the fact that the demand was greater than the supply. In

addition, stock availability was not taken into consideration. Since there was no predefined

picking schedule, they followed a just in time picking policy. The workers ability and experience
were very important in the process of picking too, since the bins contained different types of

SKUs and the worker had to read the tag of each piece to identify the wanted SKU.

Analysis

According to the case, the process of this distribution center followed many steps. Using

the floor layout in Exhibit 2, materials entered the warehouse through the folding department.

Here, the SKU was either folded or rolled according to the length of the material. After they had

been folded, they place into a trolley all mixed. This trolley would then be sent to the market

storage but before it could enter the market storage, the SKUs had to be scanned and registered at

the inbound receipts. Once scanned, the SKUs were given to the put-away team who would sort

the SKUs and place the SKUs according to the market. However, since the systems relied

entirely on the competency of the workman, it made things especially challenging when an SKU

belonged to multiple markets. For instance, out of the 30,000 pieces of an SKU, 20,000 might

belong to Market 1, 7,000 to Market 2 and the last 3,000 to Market 3. A workman who was not

aware of this distribution could have put all the 30,000 pieces in the rack for Market 1. When it

came to picking, the team would have to scramble to find the SKUs and check every market to

find it. This lead to some SKUs not being pushed out because the workers could not find that

SKU. Once the picking process was complete, the goods were moved to the packing table. The

materials were sorted and arranged according to OBDs. An OBD order had three to 50 pieces. A

packing team member entered the OBD number in the warehouse management software and

scanned the barcode of each piece. After all the pieces had been scanned, a bale number was

generated. The pieces of each bale were put inside a high density polyethylene (HDPE) bag and

the bag was sealed with a tape. Each bale was stitched by hand using hessian cloth, and manually

labeled through stencil marking. The bales were then strapped with the help of a strapping
machine, manually weighed, and sent to the dispatch unit for outdoor delivery. The packing

process was labour intensive (requiring approximately 25 persons to pack 700 bales), time

consuming and prone to errors.

In this case, the dimensions of the bins and the average of pieces were given to us.

According to Exhibits 8 and 9, the bolt bins dimensions was 36 x 36 x 29 inches. For the roll

bins the dimensions were 36 x 72 x 29 inches. The average bolt piece were 9 x 31x 2.25 inches

and the average roll piece was 6.5 x 6.5 x 65 inches. Using these dimensions we calculated the

number of pieces that we could place into each bin. We first divided the width of the bin by the

width of the bolt pieces. We took 36 inches and divided it by 9 inches per piece which gave us 4

bolt pieces per column. Then, we calculated the number of rows that we could fit within a bin.

We took 29 inches and divided it by 2.25 inches per piece which gave us approximately 12.89

bolts pieces per row. Once we got the number of columns and rows we multiplied them together

to get approximately 52 bolts per bin. We used the same procedure for the rolled pieces. We

ended up getting 5.54 per column and 4.46 per row. When multiplied them together we got

approximately 25 rolled pieces per bin. This showed us that each bolt bin could fully hold up to

52 bolt pieces per bin and each rolled bin could fully hold up to 25 rolled pieces per bin.

After we found out how much we could place in a bin, we then calculated how many bins

we would need to allocate to each market based upon the number of SKUs. Using the

information from Exhibit 6, we first calculated the average stock per SKU by dividing the total

stock in each market by the total number of SKUs in each market. We separated market 6 from

the other markets because this market held the majority of the rolled pieces as depicted in Exhibit

2 and the information that we were given in the critical facts. Once we obtained the average

stock per SKU, we proceed to find the average bin per SKU. We took the average stock per SKU
and divided it by 52 bolt pieces per bin for markets 1-11. For market 6, we took the average

stock per SKU and divided it by 25 rolled pieces per bin. After we calculated that, we multiplied

the average bin per SKU by the total stock to obtain the total bins for all SKUs in markets 1-11.

For the utilization of the bins, we divided each markets total bins of SKUs by the total of all the

bins in the warehouse and multiplied it by a hundred. This showed us a percentage of how much

we need to use in each market segment. Then we took the utilization percentages and multiplied

it by 9,967 bins that are reserved for bolt pieces to each markets bin allocation. For market 6,

however, we had to multiply it by 537 bins because market reserves only rolled pieces. This

showed us how allocation layout of the bins in each of the different markets.

Next, we needed to calculate the different meters of cloth and how many bins we would

need to assign to these varying pieces of cloth from the given allocation of bins in each market.

Using the same formulas that we used to calculate the bin allocation to each market, we were

able to find the total bins that we would need for that specific SKU in each market and were able

to find the utilization rate of those bins. The information that we used for these calculations came

from Exhibit 6 which listed the SKU and MTR of each segment. By knowing this we now know

how much to place in each bin, how many bins we need in each market, and of those bins, how

many we utilize which SKU.

Recommended Decisions

After discovering that the issue within the warehouse stemmed from different segments,

we came up with various potential solutions. First being within the picking process, we decided

to have pre-set picking schedules in place. By “pre-set” we mean that workers will be on a
schedule. They will know what items they have to pick before hand, as they will have previous

information on demand. These past reports will allow them to know and expect what to pick for

each market. The second solution we’ve come up with in regards to picking is assigning workers

to specific markets. Distinguished Market Scenarios will allow for employees to have a clear

understanding of where items should go. Rather than, employees having no direction when

transporting material from point A to point B, they will now have a clear understanding.

As far as the packing process is concerned, we will no longer be using employees to pack

up materials. An automated system will be set in place, this system will be able to gather and

wrap the materials as if an actual employee were to do it. Lastly, we have decided on a solution

in relation to the storage capacity issues. The warehouse will now not only be tracking, but

limiting the amount of items allowed to be placed per each bin. Not only will the amount of

items be limited but SKU’s as well. We have decided that it is best to design a system where only

one SKU is allowed per bin. These two changes can overall significantly affect the business for

the better.

Effectiveness of Recommended Decisions

As previously stated, we have multiple recommended decisions that can reverse these

issues within the warehouse. The reason behind having only one SKU per bin is to drastically

reduce the confusion that comes with mixing. This system needs to be as simple and organized as

possible; restricting each bin to a certain type of material can aid in doing so. Not only does the

new SKU system reduce confusion but having Distinguished Market Scenarios does as well.

Employee’s will no longer have to rely on their past experience and knowledge to know what
SKUs are matched to what market. This method of past experience can lead to both

discrepancies and differences in opinions. This type of structure allows for the ability of

employees to correctly place each SKU with it’s paired market.

To further expand upon our other stated solutions, we believe that having a certain

number of items per bin is a crucial change that needs to be made. This will noticeably reduce

overcrowding and wastage of material; which ends up being discarded or on the floor. Once we

made our calculations for maximum storage capacity for each bin we came up with specific

numbers. For the bolt pieces, we concluded that while 52 is our maximum capacity, we will set a

limit of 50. Similarly, for our rolled pieces, our maximum capacity is at 25, which we have

decided to stop at 22. The reason for this is that the thickness of each material varies. This piece

of information needs to be taken into account when making a decision on maximum limits. Since

the thicknesses do differ, it makes sense to decrease the capacity amount. This will save us from

another possible situation of overcrowding.

Execution

Execution is one of the most important steps in implementing our recommended

decisions, the first of which is setting markets. Now that we can know which markets need the

most bins we can allocate them space accordingly by changing around the rack and market

structure. Here workers along with the management team will be trained in the new market plan

layout . Workers will be instructed as to the new bin locations and how many bins are allotted to

which market. Bins will be labelled and scanned for that market only, so once a worker scans a

bin they know which market and which meter of cloth is supposed to go in there.
 The second recommendation was to have a set limit on bin items. Since we calculated we

can carry 52:25 but recommend 50:22 for safety, now the upper defined limit would be trained

into the workers as well as the system. So if a workers is assigning a SKU to a bin the machine

will stop him at the 50th bolt and assign a new bin to put the rest of the pieces.

Scenario Planning

There are many potential scenarios that can arise. Now that we have assigned bins to

specific markets, one of the problems that could arise is that demand can fluctuate. For example

if one market demand is higher than expected, we can use space from other markets for that one

month, by the time we figure an arrangement for the upcoming month. We will have monthly

market checkup to adjust market space based on forecasting. To avoid mixed skus, we will have

workers control the bins each week as inspection. We can an automated system checking item

type (b or r), SKU and capacity limit. Workers can misplace items or bins become empty of

SKUs. Weekly bins inspection will be needed to avoid emptiness of bins and also to check on

production. In case of a technological issue with the automated system. Trained workers will

make sure to respect bins regulations (50:22).

Conclusion

As conclusion, Raymond Warehouse distribution center was inefficient due to the way

they were utilizing their storage space which led to order cancellations. Opting for the storage

strategy1 SKU per bin and predefining markets by number of bins and market size, Raymond’

Warehouse distribution can be efficient.

Calculations:
36 in ÷ 9 in/per piece = 4 bolt pieces per row
29 in ÷ 2.25 in/per piece = 12.88889 ≅ 12.89 bolts pieces per column
4 x 12.89 = 51.56 ≅ 52 bolts in a bin
36 in ÷ 6.5 in/per piece = 5.53846 ≅ 5.54 roll pieces per row
29 in ÷ 6.5 in/ per piece = 4.46154 ≅ 4.46 roll pieces per column
5.54 x 4.46 = 24.70840 ≅ 25 rolls in a bin