By Rob Schneider, CEO and President, Steelwedge,
I need to get
It took a vacation
two weeks ago for me to realize that my local paper
(the Dayton Daily News) and my Wall Street Journal are
delivered by the same person.
My wife smartly
cancelled the local paper for the week we were gone.
When I returned and found the neighbor hadn't brought
any WSJ's inside, I asked my wife if she had somehow
cancelled that as well. She hadn't. To make a long
short, the local paper's delivery person simply hadn't
stopped during the week we were gone, and
ended up with a week's worth of WSJ's the following
Monday. I woke up early one day last week just to
All of which
got me thinking about logistics collaboration. Well,
I'm not sure if collaboration is really the
right word, but whatever the term should be that describes
multiple companies, even competitors, leveraging assets
and processes to meet logistics and fulfillment needs.
At one level,
the local paper and the Wall Street Journal compete
for my newspaper dollar, though I guess arguably not
too much. Nonetheless, I used to get the WSJ in the
mail – now, there's a distribution deal for the local
paper to do WSJ fulfillment.
year, FedEx announced an agreement whereby the USPS
would take over some “last mile” deliveries for the
Memphis package giant, especially in outlying areas.
FedEx delivers customer packages to the local branch
post office, which then takes them, along with the mail,
to individual homes and businesses - this despite the
fact that FedEx and the USPS compete.
Over the years,
I have heard a few plans to “re-invent” the beer industry
by developing 3PL-like capabilities that would enable
multiple brewers to ship beer to one local distribution
point, and then combine deliveries to bars, restaurants,
etc. Why should four different delivery trucks stop
at each saloon to drop off a few cases of beer?
I have also had
many discussions with CPG companies that have theorized
that eventually there will be a single truck going to
the retailer with their goods and those of their competitors.
We have some versions of this already, of course. Many
retailers manage their own inbound pools. The ES3 multi-party
distribution facility stores goods from multiple manufacturers
and delivers combined shipments to the retail store.
On the other
hand, some other attempts at related types of collaboration
never really made it. The National Transportation
(NTE), with its model for shared LTL routes, couldn't
generate enough volume, and has really evolved into
a transportation technology provider. There has also
long been interest in the full truck load arena. My
friend Bob Shagawat of Shippers Commonweath has for
years spoken and written about the potential savings
from multi-party shipper TL collaboration. I was at
a conference where Georgia Pacific spoke about their
efforts at this using Nistevo's collaboration network,
and there was substantial interest, at least in the
form of questions, from the audience. Yet, Nistevo
(or anyone else in this area) has not seemed to be
to really develop critical mass. Other software and
service providers, however, continue nibbling around
this network/collaboration area.
I realize I've
covered several only loosely-connected threads here.
More to come on this. But I think a variety of factors
(continued extreme focus on costs, still lots of
deadhead carrier miles, areas of logistics (e.g., store
that really are commodity functions) will exert pressure
so that over the next ten years we will see substantially
more of this type of cross-company activity, and
new-age service providers, technology and partnerships
enable it. Two big barriers are around information
and cost allocation. Once we solve those issues,
cultural barriers may also start to fall. Do you expect to see more cross company partnerships and
collaboration around logistics and especially transportation?
What are the barriers? How do you think those barriers
will be overcome? Let
us know your thoughts.
market took a dive this week and took our stocks
with it. Only Logility managed a small gain of
$.22 by the end of the week. Aspentech lost 16%
of its value, to make it our biggest loser on
a % basis. SAP and Peoplesoft each lost more than
$2.00 per share, our biggest dollar losers for
the week. A third of our stocks, Aspentech (up
54%), SAP (up 33.6%) and Logility (up 3.7%), are
on the plus side over the past year. All others
are in negative territory.
also took our Transportation stocks into negative
territory. Only Descartes managed a small gain,
$.07, 6.2% of it value. Our biggest loser was JB
Hunt, down $3.88, 10.1% from last week. From the
dollar perspective, FedEx ran a close second, down
$3.74 for the week. Over the past year our gainers
and losers stand at 5 each. The biggest gainer is
Yellow Roadway, up 60% for the year. The biggest
loser is Vastera, down 65.4% over that same period.
here to see performance over the past week, month,
quarter and year >>
is coming in at a rate greater than we can publish
it – thanks for your response. Our Feedback
of the Week is a short but to the point letter
on “the Bullwhip Effect Revisited”
from Jim Schultz of Texas Instruments.
find several other good letters on this topic,
as well as a couple comments on our piece about
RFID potentially being open to hacking/tampering.
If you’ve sent a comment we haven’t
published, we’ll be catching up over the
next couple of weeks.
For more complete comments
from readers, click here.
Keep the dialog going! Give us your thoughts
on this week's Supply Chain topics.
Good story in the Wall
Street Journal last week on how Toyota’s continued
growth is putting some strains on the systems and methods
that had made it the world leader in quality. It contains
some points of interest not only for those in the automotive
industry, but any interested in lean production methods.
Toyota has enjoyed tremendous success. As the article
notes, the company has “nearly doubled its revenue
in the past decade, and redefined competition in key
parts of the automotive business.” Toyota’s
profits last year exceeded those of GM and Ford combined.
But the growth, especially that driven by production
outside of Japan, has come at a cost in terms of the
quality that was key to the company’s market success
and production efficiency. It’s Georgetown, KY
plant has slipped badly in ratings from JD Power and
Associates, and quality ratings on several individual
models have also fallen from the top ranks.
What’s happened? The article suggests that with
growth, there has been a watering down in both knowledge
and belief in the famed “Toyota Production System”
that not only led to the company’s own success
but really is the foundation for much of today’s
“lean” thinking across thousands of manufacturers.
Growth has sometimes put more focus on getting cars
out the door than adherence to TPS principles. There
has been a lack of TPS experts from Japan to train North
American supervisors. Language barriers have played
a role. There has been a much higher level of turnover
in North America, both at the floor level, leading to
training issues, as well as at the executive level as
TPS experts went to Toyota rivals.
All this led executives from Japan to find that “some
hourly workers began ignoring standardized work processes
– considered one of the biggest sins inside Toyota
plants because of the impact on the consistency and
accuracy of manufacturing.”
The company is taking action, of course. More experts
from Japan have been sent to Georgetown and other plants,
and there is a new training program to develop more
non-Japanese experts. Managers have been dragged down
to the factory floor and supplier plants to gain a better
understanding of real production issues. And in the
company’s flagship operations in Toyota City in
Japan, where a new generation of TPS innovations are
delivering benefits, will soon be “exported”
to other plants. These include, for example, having
“kits” of parts delivered by suppliers and
placed into the car as it moves down the assembly line,
so that operators don’t have to look for the right
parts in kanban bins – which both takes time and
adds risk the wrong parts will be selected. Toyota calls
it “error proofing,” or “poke yoke.”
As always with WSJ articles, we can’t provide
a direct link, but please email us if you would like
a copy of this article.
Are you surprised Toyota has had some troubles maintaining
its TPS disciplines as it expands production globally?
What lessons does the Toyota experience have for other
manufacturers? Let us know your thoughts.
issue of The Journal of Business Forecasting contains
a nice article from Naresh Sadarangani and John Gallucci,
supply chain managers at Fuji Photo Film USA that contains
some advice for working with sales in the forecasting
The article contains some commonsense advice for obtaining
greater levels of sales participation in the forecasting
process. This includes listening more closely upfront
to sales issues and concerns, and quantifying the impact
of forecast inaccuracy on the company and sales. The
authors include some simple but effective models for
calculating the effects of both under and over forecasting.
Sadarangani and Gallucci then argue that perhaps the
biggest obstacle, the real or perceived impact on the
sales organization’s time, can be addressed through
focusing on a handful of “demand drivers”
for each key account. They define demand drivers as
“simply the account specific elements (such as
promotional plans, inventory strategy, and forecasts
of an account) that most impact the accuracy of the
forecasts.” [Note: the use of “forecasts”
as a demand driver for forecasts seems not right.].
Nonetheless, the authors make a good case that by focusing
on relatively few demand drivers for a market, channel
or account, the time requirements for sales are minimized,
while forecast accuracy increases. They cite a mass
merchandise customer that may have key demand drivers
related to base forecast, displays, customer inventory
strategy, and promotional events, while for an ad-driven
drug retailer they might use base forecast, opportunity
sales, ad/promotional lift, and new store openings.
When the collaboration of sales has been achieved, the
nature of the forecasting process and demand planning
technology begins to change. The key is evaluating the
forecast error coming out of the planning tool with
that of the sales team. In the authors’ experience:
tool yields the best forecast for A and B items
with strong seasonality, with little promotional
activity (modestly augment with sales input).
The planning tool is
a strong supplement to the sales input for items
with strong seasonality and high promotional activity.
The sales forecast is
actually more accurate for non-seasonal A and
B items, and most C items.
A more detailed chart illustrating this is included.
Interestingly, in two of the five cases, the authors
recommend ignoring the sales input, and in another two,
advise ignoring the forecast coming from the demand
At Fuji, this approach has led to improved forecast
accuracy at the key account/product level from 43% to
25% in two years, the impact of which is certainly in
the tens of millions of dollars in bottom line savings.
This article is not yet available on line, but send
me an email if you would like a copy.
What have you found are the keys to getting sales participation
in the demand planning process? Is the authors’
perspective on use of demand drivers the right way to
both improve forecasts and reduce the time required
from sales? Let us know your thoughts.
Full Article >>
Short article in a recent Industry
Week magazine on measuring manufacturing productivity.
It gives a brief snapshot of furniture maker La-Z-Boy,
for which “output is measured in terms of "equivalent
units." This figure is normalized for complexity,
or the amount of time typically required for assembly
of a given product. An ottoman, for example, might be
one quarter the value of a recliner or sofa. Total employment
and the number of hours worked for a given period comes
from his payroll department. The end result: equivalent
units produced per day per employee. This figure is
monitored over time, and can be compared among the eight
plants in [the] division.
Regular adjustments are key to making sure they aren't
fooling themselves, says Dave Layman, senior vice president
of operations. “Buying more components will skew
the equivalent units and drive productivity upward,
but total costs might still remain unchanged. "I
can drive that number down by taking the labor out and
purchasing it, but the end result is it doesn't turn
out," he says. “
New Balance Athletic Shoe measures productivity via
pairs produced per person per day, with headcount including
assembly and some maintenance personnel. To compensate
for the different time required to make different styles
(because some styles are easier to make than others),
the company's accounting department defines a basket
of styles that is supposed to reflect a standard amount
Meanwhile, Bridgestone/Firestone's passenger and light-truck
tire facility in Graniteville, S.C., measures plant
productivity two ways -- pounds per production per man-hour
and pounds per total number of people who work in the
The article makes the point that as the amount of direct
labor involved in production decreases for many products,
manufacturers need to look at new ways to calculate
productivity levels. It also briefly touches on what
may be the much bigger issue – the disconnect
between many plant measures that drive increased output
to improve yield or cost figures at the price of producing
inventory that no one is going to buy.
Are their changes in many manufacturing environments
that should cause many manufacturers to change the way
they measure productivity? How have you seen your company
get these productivity measures right – or wrong?
Let us know your thoughts.
I think some
progress has been made in minimizing the bullwhip effect
driven primarily by manufacturers being more in touch
with distributors, wholesalers, and retailers supply
chain inventories. However, other key factors impacting
inventory swings have not really changed much in the
past decade. In addition to the items mentioned in your
article, one other KEY contributor to the bullwhip effect
is overt management decisions to increase inventory
to capture market share during market upturns. This
factor in my opinion far out weighs all others and is
why software and technical advances have had minimal
impact on reducing the bullwhip effect.
Texas Instruments Inc.
on the Bullwhip Effect:
The bullwhip effect is only minimized when a businesses’
economies of scale match up those of the marketplace
of their vendors and customers (a random occurrence
unless a business is willing to cap sales at predetermined
production levels which leave no residual inventories,
in essence undersell your sales forecast, this will
The focus to real time tracking of information and the
push to achieve JIT have given us greater visibility
into the data that drives the "Bull Whip Effect",
but controlling the effect in total is to control the
marketplace and there are not any businesses that control
It is as simple today as it was 100 years ago …
planning, planning, and more planning - leads to enhanced
cost control - which leads to competitive edge and greater
An example would be the Wal-Mart and Costco models,
which are somewhat in control of more of their marketplace,
as they control more of the variables. Eventually though
some of their suppliers will reach a point where it
is less expensive for them to open their own retail
outlets or take on retail partners versus incurring
the excess costs that Wal-Mart may drive into their
operating bottom line by their insistence on the "Wal-Mart
way". Every model has a lifecycle and Wal-Mart's
will eventually be challenged and or modified.
Karl C. Sliwinski
The question as to whether we've made real progress
is more than fair. And, the answer is, "We haven't."
Not really. Not a fraction of what's possible.
Actually, recognition of what is called the "bullwhip
effect" goes back to the beginnings of The Beer
Game, maybe 30 years ago. Since then, there have been
any number of initiatives to balance, integrate, communicate,
etc., ultimate demand with supply chain operations.
Remember Efficient Consumer Response (ECR) 15 or so
years ago? Essentially, the same potential to take time,
cost, and delay out of the supply chain is still there.
Perhaps a bare handful of leaders and visionaries are
doing the right thing about demand communications and
ordering/replenishment, but it's still a mystery to
the mainstream. I frequently conduct a version of The
Beer Game, and the participants universally acknowledge
how the game reflects the realities of their companies'
supply chain dynamics.
Art Van Bodegraven
The Progress Group, LLC
the article "RFID Guru warns Tags vulnerable to
Tampering". It is always nice to hear such
positive information about technology from a REAL guru.
have been around and in the business for over 38 years
and have seen all types of inventory control mediums
(hand written, typed, OCR, bar coded, and now RFID)
and in all predecessors to RFID the issue of "Vulnerability"
has been there.
back one generation to bar codes, I recall a group in
the Northeast were going into KMart stores, after printing
new bar-coded labels on a portable print in their van,
and re-labeling the merchandise a fraction of the current
does not stop there! One evening I was sitting in my
car waiting for my wife to come out of the grocery store
(which at times could be hours) and the person in the
car next to me had a Transceiver picking up cell phone
only the thieves steal. I have enough faith in the major
part of the US population that this would not be the
I'm sure this fellow has some valid statements and concerns
I believe you need to be wary of using terms in your
headlines such as "guru". That gives a very
false connotation as to the statements being made. There
are no "experts" and "gurus" in
this space specific to 900mhz. Everyone is a theorist.