Senior
Business Analyst
New
Breed Companies
 |
2003 -
Present |
I am currently
working for New Breed Corporations. I was brought on board as
a technical writer for a 6-week contract. That was in October
2003. Needless to say, they liked me, wanted to keep me and brought
me on full-time. I am doing much more than technical writing.
New Breed
assigned me to two different projects:
United
States Marine Corps
Marines
do three things very well: they fight (for this I thank them very
much), they train to fight, and they hold meetings about training
and fighting. They don't manage warehouses very well, and that's
where we step in.
Although I am not an
industrial engineer, I put my general quality management and engineering
skills to work to design packaging,
develop packing standards, and set up a recovery, recycling and
retail operation for chemical protective masks.
I did a Parato analysis
on the inventory and found that 90% of the bulk of the inventory
was comprised of only 3 of the 30+ items we manage: suits, boots,
and gloves. Accordingly, I developed packaging to accommodate
these three items. The packages had to be big enough to accommodate
a reasonable number of the item, yet small enough to be manageable.
Too small a package, and there is loss of efficiency. Too large
a package and the chances of splitting lots is increased.
Another compromise
had to balance how well the items fill the box and how well the
boxes fill a pallet. I came up with two boxes that fit perfectly
on a pallet with no wasted space. For these three items, I designed
packaging with a 85% or better density efficiency. Furthermore,
the glove box packs directly into deployment containers and increases
the density by 50% over the previously used standard.
The other 10% of the
inventory bulk either fits very well in one of these boxes or
is large enough to be loaded directly on a pallet. We captured
the packing standards for these items as we brought the gear into
our first warehouse in the project.
My main area of responsibility
is the chemical protective mask. Not only did I became the subject
matter expert for the product, I also had to develop whole new
concepts for the company. New Breed is very good at warehouse
management. They take in materiel in bulk, inspect it, store it,
manage the shelf life, and ship it out in bulk. They do not do
retail operations such as individual issue and recovery of masks.
The masks require inspection,
tear down, cleaning and rebuild. We need to track not only the
stock in inventory, but also the stock that is in the hands of
the individual Marines. Quality needs to be 100%. Chemical protective
masks are life support equipment and if they fail, people can
die.
Normally when inventory
leaves a warehouse, it doesn't come back. For us, returns are
a daily occurrence. We also have to handle up to 200 customers
a day, each requiring a custom fitting and check out procedure.
We are doing this with less than 1/3 the staff that the Marine
Corps used to employ.
I also developed a
concept for bulk cleaning the masks using a commercial dishwasher.
The current process does not keep up with demand without additional
labor being applied. I designed a test, coordinated with a vendor
to conduct the tests, and ran preliminary tests. The results indicate
that the process is viable, and that formal testing should be
conducted by the Government. I did the associated financial analysis
and for the same initial investment as the current system, we
could save $90,000 in operating costs each year. The idea is currently
being considered by a Joint Commission.
I developed a flow
chart to show the end-to-end process for handling the chemical
protective mask. It is not a linear process, there are multiple
entry points to the process, multiple exit points, and depending
on the type of operation and condition of the mask, there are
various paths through the process that masks may take.
Using the inventory
transaction and inventory levels databases, as well as observations
from the field, I was able to break down each process to its most
elemental steps and assign a labor costs to each step. I was also
able to assign percentages to the paths that might be taken. The
labor rates and percentages were outlined in a flow chart and
tracked in an Excel spreadsheet that corresponds with the flow
diagram. The combination
spreadsheet / flowchart allows us to conduct "what if"
analyses to see where improvements to the process will have the
greatest impact.
In my spare time, I
develop, coordinate and conduct training. Some of the materials
I have to develop and deliver myself (or train other staff to
train them). Other materials I have to "contract out"
to the Government which means I have to find a trainer, and make
arrangements for equipment and facilities. Very often, I have
to deliver this training while our facility is still under construction
and before we actual receive the equipment.
Verizon Wireless
New Breed operates
a Reverse Logistics Center (RLC) for Verizon Wireless. Retailers
and individuals send phones back to the New Breed distribution
center to be evaluated, refurbished, reprogrammed and repackaged
for shipment.
I am involved in three
different sub-projects:
- Human Resources
Analysis
- Labor Distribution
and Performance
- Inventory Management
- Variable Process
Control
Human Resources
Analysis:
The workforce is very
dynamic and depends on hiring and dismissing temporary employees
to meet the very seasonal surges and lulls in cellular phones
returns. We also experience high turnover rates.
I developed tools to
track the current breakout of the workforce by work center, shift,
supervisor and type employee (temporary or full time). I track
conversions from temporary to full-time, and evaluate turnover
for any time period specified by the HR Manager.
I also provide tools
to the HR Manager and Supervisors that track attendance and performance
issues and disciplinary actions. Additional tools manage requisitions
for temporary employees, and audits of invoicing from the temporary
staffing agencies.
Labor
Distribution and Performance
Effective management
of the workforce depends on knowing where people are punched in,
and how much product they are producing. The output of one process
may be the input of another process, and the workforce needs to
be balanced to assure a continuous supply of product to the downstream
processes without creating an excess of work in progress (WIP).
Supervisors need to
see where their people are currently assigned, and how those assignments
affect output. They also need a tool that will allow them to see
the effects of moving people from one labor code to another to
address issues of WIP and meeting targeted goals.
I developed such a
tool. I coordinated with IT to have hourly downloads of punch
in data, use an Access Database to consolidate the data and present
it in a query, and use the query as the foundation for an Excel
spreadsheet. All the supervisor needs to do is save the spreadsheet
to the local drive, and refresh a single page (a two click operation).
Multiple reports showing the current situation are presented.
The supervisor can override critical fields such as current headcount
in a labor code and work already produced during the shift, to
show what effects these actions have on production.
Inventory Management
I developed three tools
to assist the inventory manager.
The first tool prints
labels for pallets as soon as they arrive on the dock and are
awaiting further processing into inventory. In addition to printing
the labels, the tool also tracks how many pallets arrive, when
they arrive and what type they are (for Pareto analysis later)
and how long they have been there before being broken down and
officially entered into inventory.
The next tool assists
the inventory manager with cycle counting. Existing reports tell
us where a product is located in inventory - period. Our customer
tells us when cycle counts for various SKUs are required: some
SKUs are counted every 30 days, some quarterly and some annually.
This tool helps with the scheduling of the cycle counts, and keeps
track of what SKUs have already been counted, and what SKUs are
coming due counting.
The final tool is an
aged bin report. It is important to our customer that specifically-designated
SKUs not be in our racks longer than a specified period of time.
This report identifies the location of those SKUs where the age
is coming due or past due.
Variable
Process Control
The collecting, refurbishing,
and redistribution of cellular phones is more akin to a manufacturing
process than a logistical process. However, some of the traditional
manufacturing metrics do not work for our application. In particular,
run-time charts showing statistical processes control fail.
One of the primary
reasons for the failure is that the wireless industry is still
growing, and it is also a seasonal business with swings that vary
from month to month. To complicate matters even more, we have
processes that vary by day of the week such as processes that
depend on receiving. Typically we do not receive on weekends making
Monday processing almost twice the level of other days of the
week.
The solution to this
issue was to take the growth and the seasonality out of the process
leaving daily noise behind. The remaining noise is suitable for
analysis using the traditional tools. In addition the process
is broken out by day of week. Three models are evaluated:
- By Day of Week
- “Straight
Percentage”
- Traditional Control
Lines
The software automatically
selects the correct model to use.
The signal (growth
and seasonality) is added back to the noise to compare to predicted
results.
Furthermore, there
are two versions of the model:
- The advanced version
uses MS-Excel’s solver functionality to produce a sinusoidal
model that can be used for forecasting.
- The simplified
version uses Cramer’s Rule to assign least-squares-fit
coefficients to a third degree polynomial to provide a very
accurate approximate the sinusoidal solution. This version is
100% automated and is useful if all you want to do is check
to see if recent transactions show deviations.
- Both versions plot
the measured data, signal and upper and lower control limits
and flag deviations.
|
