Professional Documents
Culture Documents
Revolutionizing
Inventory Optimization
How can advancing the way you view inventory boost
your bottom line?
Revolutionizing Inventory Optimization
Introduction
According to Supply Chain Insights as many as 75% of companies
are still trying to use Excel to optimize their inventories. Companies
relying on Excel (or other turn of the century products) struggle to
meet service level requirements while respecting financial impacts,
and have almost no chance to do it optimally. Improving inventory
optimization software remains “low hanging fruit” for many seeking
higher levels of supply chain maturity and improved supply chain
analytics.
T
he reality is the spreadsheet driven • Diverse inventory mix needs that
forecasts typically fall short of their don’t mesh well with ABC inventory
goal unless demand is 100% predictable. classification or simple rules of
However, today’s economy is never thumb
100% predictable. Here are some of • Multiple demand streams, each
the situations where a more advanced usually with different service level
approach to inventory optimization requirements
makes sense. • Global supply and demand networks
• Seasonality, replenishment
constraints
• Manufacturing constraints
• Promotional impacts
• New product introductions
Contents
04 / 12 /
How to Tell if Intermittent Demand is a Case Studies
Drag on Your Inventory Optimization
05 /
How Do I Make Intelligent
Inventory Tradeoffs
Revolutionizing Inventory Optimization
T
o understand the shortcomings of ABC to identify a service level for every individual
inventory classification, we need to SKU-Location. So a simplification is necessary
understand how it is done. Nearly all traditional and ABC classification is one way of doing it. A
inventory management applications calculate common method is a 3x3 matrix with the cost
safety stock for each individual SKU-Location value on the Y axis and order-lines on the X
combination. This requires identifying the axis - a so called “double” ABC classification.
desired service level percentage for each
The percent distribution between classes is
often based on 80% of the cost value in the A
items, 15% in the B items and 5% in the C items.
The same 80/15/5 breakdown is applied to
the number of order-lines. Because just a few
items can generate so many order-lines and
cost of sales, it usually only takes a few A items
to reach the 80% thresholds. Therefore the end
result is a matrix that looks like the following,
with a very small share classified as AA and a
majority classified as CC.
this case) is often good, and the service software offer an 8x8 ABC matrix per
levels for each class are confirmed. location. The workload to define and
From here, all of the items (or articles) in continuously maintain these matrices
each ABC class are assigned the same becomes very intense. As we said, we
service level target. In the example don’t know whether or not we have an
above, if we have 10,000 items in optimal distribution.
stock, then 5400 in the CC class will
be assigned the same service level I N V E N TO RY S E G M E N TAT I O N TO DAY
target. Then the safety stock levels
are calculated which results in a total Is there another way to address safety
inventory investment. stock computing in 2016? The answer is,
of course, yes. A more modern approach
Let us think about this for a moment. exists.
Inventory investment is a consequence
of each ABC classes’ service level. Could Traditional ABC classification is based on
we have chosen other service levels an operational or logistics perspective.
for the classes and still reach 95.5%? There is rarely any connection to sales
Of course! There are a large number of and marketing or the companies’
combinations that could result yield the customer needs.
same result.
Inventory optimization instead looks at
How do we then know that the the product range and the business. This
distribution we chose is the most difference is possible thanks to the use
optimal one, achieving a minimum stock of "service class". Examples of service
investment? The answer is that we don’t classes can be ”accessories”, ”items with
know. That is why this method is called a high margin”, ”own-brands”, ”high end
”inventory management” rather than brands”, and ”critical spare parts”, to
”inventory optimization”. name a few. This type of categorization
In this example, the process for one is much more relevant to sales and
echelon is described. Many supply marketing, who often have very little or
chains are more complex, with several no understanding of ABC classification.
connected echelons such as central, Every service classification contains
regional and local inventories. Also, items from several ABC classifications
some traditional inventory management (according to the old method) which is
2 / toolsgroup.com
Revolutionizing Inventory Optimization
3 / toolsgroup.com
Revolutionizing Inventory Optimization
4 / toolsgroup.com
Revolutionizing Inventory Optimization
L
uckily there is a solution: modeling behavior. The outcome is a value within
demand and uncertainty and using that a range, and each value has a certain
knowledge to make intelligent inventory probability of occurring. This leads to higher
tradeoffs that enable high service-levels and service levels and revenue growth. Here’s how.
customer satisfaction. Traditional forecasting
systems can't do this because they are D R I V I N G D OW N TO T H E D E TA I LS
“deterministic”—their internal processes view
all data as exact. They take exact values Demand modeling breaks data down to make
as input and they output exact values, so the baseline demand as detailed as possible.
the forecasting calculation is oblivious to The most common granularity is the individual
the uncertain nature of the demand. Any sales order-line, by item and by ship-to
deviation in demand, no matter how normal, is location daily. This data is readily available at
considered as error. companies that maintain detailed demand
histories in the form of individual line-orders.
Demand modeling works differently; everything They know if monthly or weekly demand for 48
is “stochastic” (probabilistic). Stochastic cases was generated by one very large order
modeling systems train their sights on a for 48 cases or by 12 smaller orders with an
more accurate forecast, but by modeling average size of 4.
probabilities and then factoring in random
5 / toolsgroup.com
Revolutionizing Inventory Optimization
B E T T E R D E M A N D DATA I N F O R MS
S M A RT E R I N V E N TO RY T R A D E O F FS
6 / toolsgroup.com
Revolutionizing Inventory Optimization
being 1 unit or 10 units, or some number in includes 4 factories, 23 production lines, and a
between, can be far more important than the distribution center. Products reach market via
actual forecast. four channels—retail, food service, industrial,
and international.
Stock-to-service modeling can then drive
the stock calculation and inventory tradeoffs. Like many food manufacturers, Findus favored
To achieve a target service level requires service levels to meet the tough demands
generating a stock-to-service curve that of customers. This meant holding artificially
specifies how much inventory is required for high safety stocks to ensure availability, which
each level of service. didn’t sit well with board members. But a
new demand modeling system let planners
This is done by incorporating a statistical update and control safety stocks levels every
description of all the demand streams and all month, rather than every six months as before.
the replenishment parameters. These models It also allowed the team to advance from a
take into account the expected demand “one-size-fits-all” channel strategy to treating
behavior and factors like inventory control each channel differently, according to its
policy, minimum and incremental lot-sizes, unique demand patterns, volumes, and other
lead-times and lead-time variability, review properties.
periods, etc.
Findus topped its original goal to cut inventory
These systems deliver the granular knowledge
to make intelligent inventory tradeoffs that
deliver service levels that support revenue
growth. Inventory has costs—working capital,
stock management, space, insurance, stock
loss, expiration, obsolescence, damage, and
markdowns. Yet it is crucial to service levels
and managing unforeseen events.
A N E X A M P L E O F S E RV I C E L E V E LS
D R I V I N G G R OW T H
7 / toolsgroup.com
Revolutionizing Inventory Optimization
by more than 10%, reducing working capital SKU demand variability at the heart of the
requirements by more than SKR 20 million ($3 inventory issue. It sets targets at an SKU, not
million). Once the results demonstrated to the aggregate level, for each stocking location.
board that service didn’t have to come at the Because it precisely targets and continuously
expense of capital, the supply chain team optimizes inventory, firms don’t blindly
switched from its mandate to free working overstock, but buffer only where there is a
capital to prioritizing service levels. They then likelihood of demand and sales.
increased service levels significantly, especially
in the retail channel, to 98%. The result is that excellent customer service
levels up to 99+ percent are now achievable—
D E M A N D M O D E L I N G A N D I N C R E AS E D with a huge payback. Even a modest increase
S E RV I C E L E V E LS L E A D TO G R OW T H in fill rate generates fewer lost sales, higher
market share, more margin contribution, and
Demand uncertainty is best managed the ability to satisfy even the most demanding
using planning tools that employ adaptive, customers. By understanding each SKU’s
probabilistic algorithms. They deal with stock-to-service relationship, advanced
demand volatility and demanding response demand-driven modeling and inventory
times, particularly as caused by online optimization can help you reach your service
and multi-channel markets. Their adaptive objectives to power revenue growth.
modeling includes a high level of automation,
so planners need to intercede only for
exceptions that fall outside the boundary
limits of statistical uncertainty.
8 / toolsgroup.com
Revolutionizing Inventory Optimization
C
ompanies have turned to inventory • What capabilities are unique to your
optimization solutions as a way to set industry?
precise stocking levels for thousands of stock
keeping units (SKUs) in their supply chain. For B R E A DT H O F F U N CT I O N A L I T Y
companies running complex supply chains.
Before contacting providers with an RFI, Providers should provide multi-echelon
Nucleus first recommends that you determine (MEIO) solutions as standard functionality,
what the primary business reason for buying thus allowing companies to adjust inventory
the tool is. holistically across multiple locations in the
supply chain. The solutions should also employ
a stochastic model for inventory calculations,
• Is the tool needed to reduce inventory meaning that the application not only
to free up working capital? Or does the identifies the most likely outcome (the forecast)
company need to ensure that it has the but also the implications of possible upper
right amounts of inventory and inventory and lower limits of demand fluctuations.
mix on hand to support sales and drive
revenue? To be effective the tool must be able to model
inventory at the SKU level and by location.
• Does the company have a problem with Granularity at the item level is critical to
maintaining stock to address growing determine the precise inventory holdings
volatility in the market? by location to meet anticipated demand.
Keep in mind that IO tools are really a form
• Does the company want to keep the of predictive analytics, and the best-in-class
right parts in stock to reduce expediting tools are the ones that do the most accurate
costs and provide a more level signal to job of predicting the amount and type of
production scheduling? stock to carry at each location. Generating
high-level predictions and then smoothing
9 / toolsgroup.com
Revolutionizing Inventory Optimization
them to the individual SKU-location level will inventory models are rather complex and
not get the job done right. not very intuitive, especially when optimizing
across either a wide range of products or
Since the key to IO is modeling the supply upstream and downstream in the supply chain
chain accurately, if that’s not done right, (e.g., postponement). It’s important that the
a company won’t get favorable results, no IO tool provide an automated solution so that
matter how many features and functions can the planners can use the output of the tool to
be found in the tool. Supply chain executives focus on exceptions from plan or problems. A
should not get caught up in the old way of sustainable solution should allow the business
buying transaction-based software – with to scale up, handling additional it as the
long detailed lists of feature and function business grows or changes, or the supply
requirements along with vendor responses. chain operation becomes more complex. In
short, the solution should automatically and
CO N D U CT A P R O O F O F CO N C E PT continuously adjust a variety of conditions
and changes without constant manual
Before making a firm commitment for a intervention.
solution, a company should test drive the
software. A Proof-of-Concept (POC), involves E AS E O F U S E
providing actual data to the vendor and
asking the vendor to validate the software’s The selected solution should be easy for
ability to build a working model of its solution. supply chain planners to use in their daily jobs.
The data selected could be for a geography, a Although many of tools offer a high degree of
product line or some representative portion of functionality, planners still need to understand
your business. Providers should be given actual the application and the results. For instance;
company data for the trial. Take a significant
portion of items from the company’s inventory • They need to be able to define their
and ask the vendor to run its model to set “service policy” (target service levels for
inventory levels. Then compare the vendor’s classes such as high margin products),
test scenario with what actually happened. so the IO system can translate service
The test should be substantial enough to targets into inventory targets.
verify the sustainability of the process and the
scalability of the solution. A successful test • Leading Providers should offer a graphical
should offer proof that the solution can indeed user interface (GUI) that allows a user
improve service levels and minimize stock, to grasp visually the intent of “what if”
thus offering a clear indication as to whether scenario settings.
the company will get a payback from the
investment. NOTE: The proof of the test should • The tool should come with dashboard
help avoid a “feature and functionality” bake- analytics, a grouping of widgets in a
off – where the bells and whistles supersede console that allows the user to spot key
the actually functionality of the proposed metrics at a glance.
solution.
Ultimately a supply chain planner doesn’t
AU TO M AT I O N O F P R O C E SS / R E D U C I N G need to manage the minute by minute inner
M A N UA L TAS KS workings of the analytics engine but does
need to be alerted to unusual conditions that
Consider how automated the solution will require intervention.
be once it is set up and running. Optimized
10 / toolsgroup.com
Revolutionizing Inventory Optimization
11 / toolsgroup.com
Inventory Optimization Success Stories
12 / toolsgroup.com
LENNOX
Lennox International is a leading provider of climate control solutions for heating,
air conditioning, and refrigeration markets around the world. The Residential
HVAC business unit represents about 42% of $5 billion total company sales.
CHALLENGE
Lennox Residential Heating and Cooling faced the challenge of managing an ambitious North American distribution network enlargement
while simultaneously transitioning to a hub-and-spoke model with 55 shipping and 161 selling locations.
Rebecca Roberts, Director of Supply Chain Planning, realized that the business scenario called for a flexible, self-adjusting system with
comprehensive statistical forecasting that could set stocking targets at an SKU-Location level based on demand variability, service levels,
costs and other parameters.
SOLUTION
Lennox implemented ToolsGroup’s SO99+ solution to dynamically rationalize the inventory mix and create an operational plan that sets
inventory stocking targets and balances service levels with inventory cost. The system allows Lennox to reliably model both seasonality and
variable demand patterns. Network inventories are rebalanced by creating a dynamic optimal mix of inventory and service levels down to
the store level. Lennox can also set global service policies by group or category and then the system automatically calculates individualized
service levels for each SKU-Location.
The solution is integrated to Lennox’s SAP APO platform. Deloitte Consulting provided integration services and addressed key strategy,
technology, and change management issues.
RESULT
• Improved service levels by 16%
13 / toolsgroup.com
PIAGGIO
As Europe’s largest manufacturer of two-wheeled vehicles, Piaggio is known
throughout the world for superior technology and design. Since 1946, Piaggio’s
stylish and fuel-efficient brands of motor scooters (including the famous Vespa)
have been the epitome of practical indulgence.
CHALLENGE
Piaggio USA’s new Director of Spare Parts, Marco Ciccolini, found himself working extra long hours and even weekends to field numerous
complaints from dealerships in the wake of abundant backorders. He also found that he was dealing with a high level of dead stock
resulting from an inability to sense regular versus specialised demand. His challenges were compounded by managing three different
product lines with a very heterogeneous aftermarket part mix. And despite experiencing service levels that had dipped as low as 50%, his
organization was relying far too much on expediting shipments via air freight.
SOLUTION
Within a few months, the process for fulfilling spare parts orders was quickly revamped. The SAP R/3 system now feeds data into a parts
database and inventory file. SO99+ takes this information, applies analytic models, and generates replenishment proposals. These proposals
become the basis for parts orders, which are finalized in SAP. The overall process both identifies the inventory needed to fulfill orders and
models in-transit and safety stock to minimize inventory while maintaining high service levels.
RESULT
• Five-day service levels across Piaggio USA’s brands leapt into the 90% range, sometimes exceeding 95%, and most important,
remained highly consistent thereafter
Despite the significant upswing in service level, the company’s inventory turns actually improved slightly. Going forward, Piaggio USA’s goal is
to maintain these high service levels while further fine tuning the system to increase turns.
14 / toolsgroup.com
MOHAWK
Founded in 1931, Mohawk is a fourth-generation, family-owned and operated
business based in Cohoes, New York, with global sales and operations located
throughout North America, Europe and Asia. Mohawk is North America’s largest
privately-owned manufacturer of fine papers and envelopes which are preferred
for commercial and digital printing, photo specialties, social stationery and high-
end direct mail.
CHALLENGE
The fine paper industry is unrecognizable from 20 years ago. The internet, globalization and changing customer demand have transformed
it from being a traditional, high-volume and personal business to one where it is now possible for a customer to buy a single sheet of paper
online. Naturally, these changes have resulted in market consolidation. Mohawk Paper Mills acquired the Fine Papers division of International
Paper, followed by an acquisition of the brand assets of Smart Paper and recently Crane Stationery, making it North America’s largest
SOLUTION
Mohawk’s SO99+ deployment plays an essential role in Mohawk’s ongoing transformation program. It supports collaborative goal-setting,
negotiation and decision-making by giving all players in the extended supply chain a transparent, shared view of data that they can trust.
Mohawk continues to fine-tune its existing S&OP process to optimize inventory, which is powered by ToolsGroup’s SO99+. This defines
inventory targets for each stock item at each location across the network and performs forecasting, service-level planning, and inventory
mix optimization. It also handles requirements planning, resupply of regional (level two) DCs and passes manufacturing replenishment
requirements to their production scheduling system.
Key to bringing Mohawk’s extended supply chain into the transformation program was establishing a new data management team. This
team provides specialist data hygiene, modeling and analytics skills to support the distributors, which aren’t set up to bring this expertise
RESULT
• Transformed to a supplier collaboration culture based on data, trust and shared objectives
• Customer-service levels increased from 85% to 93% in just six months with significantly fewer costly rush orders
• Global inventory lowered by 25%, with 8% sales growth and over 40% improvement in inventory turns
15 / toolsgroup.com
Book a Demo
To learn how you can take control of your supply chain,
call the ToolsGroup team at 617-600-7522 or visit
https://knowledge.toolsgroup.com/contact-toolsgroup