Measuring Forecast Accuracy

Forecast

Supply Chain Forecast Accuracy is usually measured with Mean Absolute Percent Error or MAPE, the average of percentage errors. But there are several other metrics to consider. Here’s an example …

Continue reading “Measuring Forecast Accuracy”

Kanban and Sustainability

LuizDear Larry, Andrew and Guanair; How are you doing ? I would like to share with you that We Won the Whirlpool award Sustainability-2013 , among 536 suppliers and 55 projects, With our project about plastic returnable box saving industrial waste (avoid 75000 cardboard boxes /2500 wood pallets /50 km of plastic wrap /40 ton of contaminated waste in the last 5 years) and inventory reduction from 20 to 8 days of course. We are doing our part for a sustainable world. Thank you very much for the support.

Merry Christmas and happy new year!

Luiz Antonio Basso
Manufacturing Manager
Associated Spring, Barnes Group
Campinas, Sao Paulo, Brazil

 

kanban1 kanban2

See kanban simulation

Demand Segmentation – one size fits none

one_size

We know better than to try to force fit strategies yet time and again we find businesses planning and managing their processes with a ‘one size fits all’ approach.  Hospital services, insurance underwriting, pharmaceutical production, consumer electronics assembly, integrated supply chain services all struggle with finding the right balance of specialization and common process.  At the extremes there is the chaos of infinite flexibility where every order, customer, patient is an exception to the other extreme where the ‘cats and dogs’ are blocking traffic on the highway.

Simply sorting process demand (orders, service requests, projects, patients) by volume or frequency isn’t good enough.

Adding the dimension of demand variation is an improvement, but optimizing inventory, process performance, supply chain doesn’t end with a 2×2 graph of Volume vs. Coefficient of Variation.  But before we tackle total cost, margin analysis, supply chain policies let’s go over the fundamentals.

ABC Analysis

Gather up representative data for orders, arrivals, requests – whatever you have that represents demand on your process, business, factory, enterprise.  In HR this might be recruiting requests, in product development an inventory of active projects, in a factory grab customer orders. [ABC Analysis: how to]

ABCNow take the history and summarize product, or service and make a PQ Product vs. Quantity Pareto histogram like this …

Sort the product or services from high to low, left to right.  The highest volume activities want to be streamlined, maybe even have dedicated processes and resources.  Call these A items the ‘highway’.  The lowest volume of orders or frequency of service items, the C items may need to be handled on a case-by-case basis.  If you don’t then you’ll have a ‘mixed model’ process where the A’s and C’s get in each others way.

Usually:

–A’s  = dedicated cell or line or department
–B’s  = mixed model
–C’s  = “craft” or specialty work cell, or premium service department

Cv ExamplesDemand Variation

ABC PQ Analysis only goes so far and as previously discussed [1] [2] [3] doesn’t consider that while most high volume items are usually stable and predictable some of the high volume items may be unpredictable and have erratic demand, and many of the low volume items will be well-behaved and very repeatable.
If you made a time series graph for every product or service you would be able to ‘eye ball’ the period to period variation in demand and perhaps make some judgements on categorizing similar profiles.
Here are three examples from a recent pharmaceutical demand analysis.
The first product has fairly stable weekly average demand.  A weekly Coefficient of Variation of 0.33 might be considered a little high by some.  The spike at around week 52 is a year-end effect, so overall this product is well-behaved.
The second product is half the volume of the first visually more ‘noisy’.  A weekly Cv indicates that this product would be a good candidate for pack-to-order, otherwise significant inventory would be need to be able to cover the spikes.
Product 3 is very unstable.  Any guess on how much inventory to carry, raw material to buy, capacity to reserve, or lead time to promise is certain to be wrong.  This product would be a good candidate for make-to-order, move to a low volume – high mix business unit (and price and staff accordingly), or possibly discontinue.
Any attempt to try to efficiently run these three products through the same forecasting, planning, scheduling, production, distribution process is bound to be flawed.  The lead times, lot sizes, equipment assignment, and other planning parameters should all consider the item’s demand pattern.  Where you find high Cv ‘cats and dogs’ items coexisting with low Cv  ‘steady eddy’ products you need to seriously consider applying demand segmentation and not trying to make one size fit all.

Product Lifecycle

Products and service have a lifecycle; usually new products volume increases and the initial demand is spotty but stabilizes as the product matures.  Then new products compete and demand declines and becomes irregular for the first generation.

Lifecycle

Growth

Prime

Retire

 

Demand Profile

Customer demand can be described, quantified.  Here’s how …
1. Pick product, product family, or business unit of interest
2. Determine an appropriate time unit – hourly, daily, weekly, monthly
3. Gather the true customer demand as best you can.  Be careful about using promise dates instead of requested dates, and be doubly cautious of schedules which are often smoothed, filtered, or otherwise manipulated.
4. Create the graph or time series plot
5. Then calculate some simple descriptive statistics and interpret
An example…
TimeSeries
Descriptive Statistics
Then use Excel Data/Data Analysis/Descriptive Statistics and interpret the results.  In this example mean, median, mode are all close to 17 per day.
The range is 49 (max-min) and +/- 3 standard deviations is 34, or -16 to 50 (17+/-3*11)
(or zero to 50 since we don’t usually have negative customer orders, and we don’t usually worry if demand is lower than the mean, it’s when demand is far greater than the mean that things get interesting.)
The standard deviation is less than the mean, so the demand pattern is somewhat stable.
This demand is a bit skewed to the right, but only slightly;
confidence limit estimate =2*SQRT(6/N) = 0.94
A confidence limit estimate for kurtosis is 2*SQRT(24/N) = 1.89.
Since 1.45 is within +/- 1.89, we can say that the data is marginally normal.
The confidence limit is smaller than the mean, so the mean is reliable.

 

Process Model Selection

Method by Segment

 

Other considerations

  1. Margin Contribution & Cost to Serve may be more insightful than just looking at demand quantities or sales dollars.
  2. Products and services may have different demand, inventory, customer replenishment, supplier replenishment, sourcing and total landed cost, and other policies.
  3. Managing these policies in a dynamic business can be a challenge; so look toward optimization automation to stay ahead having the right strategies for the right products instead of one size fits none.

 

 

Lean Forecasting

Why can’t Sales give us a decent forecast?  Paraphrasing George E.P. Box, “All forecasts are wrong, some are useful.”  If forecasts are wrong, then why forecast?  Supply chain lead time is often much longer than the customer is willing to wait, and so the business carries inventory somewhere, and inventory is a waste.  The Lean approach to crush lead time and make-to-order isn’t always possible and so we may need to build on speculation, or at the very least buy raw materials.  Even after making customer consumption visible to all, reducing lead time, creating flow, using pull systems, increasing velocity and reducing variation, and establishing a multi-echelon collaborative and process disciplined environment you may still need to ‘predict’ future demand volume, product mix, and/or seasonality. So whether you want to or not, or have a formal forecasting method or not, somewhere in your purchasing and planning world someone is anticipating what the customers may want and is placing supply orders or adjusting inventory levels or rearranging workloads. Lean Forecasting Principles:

  1. Measure forecast accuracy, seek root cause for inaccuracies, and strive for continuous improvement of the forecasting process
  2. Aggregate (family) forecasts are always more accurate than individual SKU’s – plan families, build SKU’s on pull
  3. The ‘best’ forecast is a customer order, the next best is to replenish what the customer just consumed
  4. Forecast accuracy usually degrades the further out – try to match the forecast period to the supply lead time
  5. Relentlessly seek to crush supply lead time and variation (mix and volume), and eliminate the need for near term forecasts
  6. Drive waste out of the forecasting process

A collection of current thinking on Lean Forecasting:

Lean and Forecasting from Michael Balle’s Gemba Coach Column

What Happened to Materials Management and Forecasting Part II by Bill Waddell at Evolving Excellence

Building a Lean Fulfillment Stream by Martichenko and von Grabe   .

Safety Stock

 

Over time, inventory goes up and down.  As customer order shipments are made inventory transactions subtract the ship quantity from the quantity on hand.  When production is made or goods received into the warehouse the quantity on hand goes up.  This up/down cycle can be thought of as a ‘saw tooth’.  Except it’s never as neat and tidy as this.

Cycle stock can be calculated as ‘Demand during Lead Time’, or if we make or buy product once a period I should have at least enough to last until the next order arrives.  One way to do this is to first determine the replenishment time (RT) or the time between orders.  For high volume products we may make the product every month, so the RT would be 30 days or 4 weeks or 1 month.

Next we need to know the consumption, sales forecast, or historical shipments and then calculate the average demand during the replenishment period.

 

AD = average demand, RT = replenishment time, or campaign interval, or period between receipts.

In this example we receive 400 every period. Our stock on hand goes from 400 down to zero just as the next resupply arrives.  The average inventory isn’t 400 or 0, it’s 200.  When there are many products being received and issued every day some are near max, some are near zero, but the average for all products is ½ (max – min).

But this is still too simple.  Consumption (issue from stock, or shipments) is never exactly smooth, sometime higher or lower than planned.  Sometime resupply is earlier or later than planned or in different quantity than planned.  So we have two types of supply & demand variation to think about.  We need safety stock and safety lead time to protect our customers from variation in supply and demand.

 

In this example we have 100 in safety stock.  If everything were perfect inventory would cycle between 500 and 100.  We keep 100 just in case production is late or customer demand is higher than expected.

Demand can come in faster or in greater quantity than we had planned (red dashed line).  If we can’t get the product resupplied sooner we have a period of stock out or back order.

 

With safety stock we can try to prevent back orders or stock outs.

We dip in to safety stock when resupply is late.

Putting this all together we get …

 

Example:

Demand Safety Stock can be calculated, but first we need to look at some statistics.  Here is graph of weekly shipments for  CARILATUDE 350MG for the period of 1/1/2012 to 12/31/2012.

 

Here are some descriptive statistics:

Total shipped = 1,066,772
Weekly Average  = 20,449
Max Week = 61,776
Min Week = 6,360
Weekly Standard Deviation = 8,883
Coefficient of Variation = Std Dev/Average = 0.43

How much inventory should we keep for this product?  Do we want to have enough to cover the two big spikes or is it OK to go on back order for a week or two?  Before we can answer these questions we need to know a few things.

  • How often do we receive this product in to stock?  If we make the product once a month, then we need to have enough in stock to last for a month plus any uptick in demand and some more to cover if production is late on the next run.
  • What service level does management want to achieve?  You would need infinite inventory to cover all possible customer demand and supply issues.

We don’t have enough money or warehouse space to hold all the stock we might want.  So we have to accept some level of stock out or back order.  Where we carry more stock than we need we have consumed raw materials, production capacity, machine time, labor that could have been used to make other products.  When we don’t have enough stock on hand we are often forced to change the schedule and make product earlier than planned, and this means other products get pushed out.  The trick is to set just the right level of target inventory with not too much safety stock.

 

To set Safety Stock we can consider three types of additional inventory based on the source of variation:

Demand Safety – customer demand can come in greater volume than expected and deplete inventory before the next planned replenishment (raw material or production run).

Supply Leadtime – production can be late due to machine breakdown, labor availability, scheduling changes, lost batch, or get hung up for a variety of reasons

Murphy – unpredictable circumstances that management might want to hedge.

There is also seasonality and planned stock builds, but we aren’t addressing those here.

Demand Safety is based on service level and demand standard deviation.

The safety factor is based on the service level management is trying to achieve.

Service Level % Safety Factor
60 0.26
70 0.52
75 0.77
80 0.84
85 1.04
90 1.28
92.5 1.44
95 1.65
97.5 1.96
98 2.05
99 2.33
99.5 2.58
99.9 3.08

 

When we calculate, for each product, the weekly shipment historical margin contribution average, standard deviations, and Cv and then graph the average vs Cv for each product we get the following Demand Segmentation graph …

Here we see that there are many product with low volume that are very unpredictable (Cv > 1.0)  unfortunately we may have to keep stock for these products.  Because the history shows that shipments are erratic we may have to carry a lot of inventory to be able to cover the unpredictable demand.  Fortunately most of the high volume products are predictable, but some are on the edge of being noisy and so we may have to carry a large amount of safety stock for these too.

Material Classification Logic

Bruce asked “Do any of you have any standard material classification logic that you have used.  I’m working with a client to classify all of the parts they purchase.  In the past I’ve used some standards from NAPM but I can’t seem to find them now on the ISM site.”

I thought for a moment and went browsing and here’s what I can up with:

Have any better ideas, lets us know!

Square Root Law – inventory in multiple locations

Got asked what would happen to inventory when the number of stocking locations change.  I thought for a minute and remembered a quick rule of thumb … The Inventory Square Root Law states that …

Average inventory increases proportionally  to the square root of the number of locations in which inventory is held.

X2 = (X1) * √ (n2/n1)

n1 = number of existing facilities
n2 = number of future facilities
X1 = existing inventory
X2 = future inventory

Here’s an example:

Current inventory is 4000 units, 2 facilities grow to 8.  Using the square root law the future inventory = (4000) * √ (8/2) = 8000 units.

 

* The Square Root Law was mathematically proven by D.H. Maister in his 1975, International Journal of Physical Distribution article entitled “Centralization of Inventories and The Square Root Law.”

 

Safety Stock to Bridge the Forecast-Accuracy Gap

(From David McPhetrige, founder of TopDown Lean Systems, LLC, providing correct, comprehensive, multi-attribute safety stock analysis, http://topdownleansystems.com.)

Surveys indicate that even world-class companies have average forecast accuracy in only the high 70%’s, especially at the SKU or component level that makes or breaks service levels and financial performance. Contributors to forecast inaccuracy include factors that can be addressed through best practices, and factors that will always be unpredictable:

  • Chronic bias, which can be minimized with effective Sales/Inventory/Operations Planning, or SIOP. (It’s a safe bet that world-class businesses are already doing this, and they are still achieving only high 70%’s.)
  • Unforeseen and unforeseeable special causes, such as natural disasters.

The reality is that forecasting can be only so accurate, in part because it predicts the timing and magnitude of only three types of variation:

  1. Trend
  2. Seasonality
  3. Certain foreseeable special causes, such as promotions

A forecast can predict the timing and magnitude of these variations only so well. Realistically, then, many businesses – and even world-class companies – are left with a significant forecast-accuracy gap that, if not bridged, compromises target fill rates, inventory performance and financial goals.

So first, how do you bridge this gap today? Obviously, you strive continuously to improve your forecasting accuracy, and you always will. That said, though, here are some bridging “techniques” that I’ve observed and even been part of, and likely you can think of some that I’ve missed:

  • Mind-set: Complain about, but begrudgingly resign yourself to, suboptimal inventory and fill-rate performance, and related costs and risks
  • P&L: Increase the budget for airfreight, expediters or other expediting costs and expenses
  • Balance Sheet: Increase safety-stock levels
  • Obsolete/Slow-moving: Increase efforts to return excess materials to suppliers

Some of these techniques can and do help achieve service-level targets. But at a minimum, they rarely help, and may in fact hurt, financial performance. Fortunately, a financially-beneficial service-level bridge from forecast to reality is right there in your data! How so?

Well, there is still variation that’s left over after bias, unforeseeable special causes and the three variations with predictable timing: Common-cause random variation in demand (or usage, for components) and replenishment lead time.

The good news is that it is possible to quantify the magnitude of random variation by analyzing historical demand and/or lead-time data. Of course, the timing of random variation is unpredictable (that’s what makes it random), and this means that common-cause random variation must be addressed with properly-determined safety stock.

At this point, you may be saying, “I’m already doing that. I use a statistical technique to determine my safety stock levels.” Forgive the blunt and clichéd reply, but – how’s that working for you? Honestly, is your technique consistently achieving your fill-rate targets on an item-by-item basis? Or does it often seem to put too much inventory in place? And in many cases, does it put too little inventory in place, and you have to subjectively override or increase the calculated safety-stock level?

The fact that your safety-stock calculation is unreliable does not mean that there is no statistical solution. What it does mean, however, is that you must

  1. Properly identify and isolate the common-cause random variation from all the other variations, and
  2. Use a correct, comprehensive statistical safety-stock approach that includes not just common-cause random variations in demand and lead time, but all six factors that affect safety stock, service levels, inventory performance and expediting.

We aren’t talking about a spreadsheet, or perhaps an unused ERP feature. We’re talking about outside expertise in safety-stock analysis.

A recent CSCO Insights executive brief (from Supply Chain Digest and Cognizant) entitled “Five Strategies for Improving Inventory Management Across Complex Supply Chain Networks” recommends this as the first of five strategies: “Get Much More Granular with Safety Stock Management.” (http://www.scdigest.com/assets/reps/exec_brief_network_inventories.pdf.)

This brief advises using “many more attributes associated with each SKU.” The result of this expanded effort is compelling: “The greater the [safety-stock] precision a company will have.” The CSCO Insights authors advise that to do this “obviously requires a lot more work,” and that this increased analysis offers “rich dividends.” The brief concludes that pursuing the strategy of a “higher level of safety stock management” may require a “relatively uncommon skill set” that may best be provided by outside expertise.

Total Flow Management

Euclides A. Coimbra and his associates at the Kaizen Institute have created a wonderful and detailed work on the application of continuous improvement to supply chains.  Here is a full exploration and application of lean from end to end of the extended value stream that they call Total Flow Management.  Two thumbs up!

Some of the graphics look to once have been powerpoint and when reproduced are to small and grainy to be able to read.  There isn’t an index so finding topics is limited to the table of contents.  The book is hard bound, and printed on good paper.

Some of the vocabulary is odd; “border of line” might be better said as” interface” or “borderline”.

Economic Order Quantity, or as referred to in this book, Wilson’s Formula, is treated in a refreshing way.

We can say that Wilson’s formula still applies today.  The only problem is when people assume that changeover time (or, generally speaking, ordering cost) is rigid and cannot be reduced.  Many people don’t think to do Wilson’s calculations because they are still misled by two strong paradigms: flow at any cost and efficiency at any cost.

The ‘flow at any cost’ paradigm is a rising paradigm that is currently gaining in popularity.  People hear about the wonderful Toyota Production System (TPS) and start to increase the flow by reducing the batch sizes blindly, without looking at Wilson’s formula.  What happens is that the CAPEX requirements explode, because the small batch sizes together with big changeover times decrease efficiency.  The result is that flow is indeed achieved – but at the expense of capital expenditure, not by internally reducing the changeover time and increasing equipment flexibility.  You can see this effect in many rich companies that are implementing Lean manufacturing and the TPS.

For a more in depth review check out Jon Miller’s posting on Gemba Panta Rei,
Review of Total Flow Management by Euclides Coimbra.

Replenishment Strategies

Determining an appropriate production model starts with Demand Profile and Demand Segmentation.  High volume low variability items, and low volume high variability items behave very differently.  How to decide if a particular product is a candidate for a one piece flow cell or a craftsmen job bench?  Look to the coefficient of variation for a clue.

Demand Segmentation - Volume vs Variability

 

Type 1 – Rate-base or Just-in-time

  • forecasting of the flow rate or takt time
  • RCCP – rough  cut capacity planning to monitor impact of mix and volume on pace maker operation
  • produce to rate (or TAKT) vs discrete order or customer pull
  • demand flow vs time-phased requirements planning
  • maintain flow priority and timing
  • no detailed Capacity Requirements Planning required
  • no or minimal shop order launch or inventory transactions
  • highly visual and standardized shop floor control
  • “one-piece” flow, zero inventory, standard WIP – work-in-process
  • seamless flow/pull of material
  • Dynamic cycle time (Little’s Law)

Type 2 – Pull

  • combination of discrete forecasting and/or demand rate-based forecasting
  • MRP planning — pull Kanban, Heijunka visual shop floor control
  • RCCP, but no detailed CRP
  • flat Bills Of Materials
  • more cellular manufacturing
  • point-of-use vs. central stores
  • inventory is strategic: standard inventory, time-based replenishment, pull based on consumption vs. push based on demand
  • based on statistically balanced rate, build to level-loaded demand with calculated standard inventory buffers

Type 3 – Push or Job Shop Discrete

  • discrete requirements planning (firm orders and long range forecast)
  • Rough Cut Capacity Plan
  • time phasing of requirements
  • application of order policies: lead time, safety stock & time
  • Capacity Requirements Planning
  • MRP shop order launch & order maintenance (message filters and “noise management”)
  • ability to aggregate disparate requirements across multiple products by work center, supplier, product
  • central stores of inventory
  • multi-level inventory: stores, pick, kit, move, queue
  • batch processing
  • demand leveling difficult and uneconomical

Demand Profile

Maslow’s hammer, or a golden hammer is an over-reliance on a familiar tool; as Abraham Maslow said in 1966 in A Psychology of Science, “It is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.”  So, must every product in every business segment be set up in a one piece flow cell? Or put on kanban with an heijunka to smooth demand? Or run on a rate-based assembly line? Certainly not!  One size rarely fits all.  But how to know which techniques make sense?

One place to start is to look at customer demand. All lean practitioners know about Takt Time, or the customer drum beat, that is used to match the pace of an operation with customer demand.  Takt Time is calculated as Available Time/Demand, and is by definition a ratio of averages.  But customer demand is anything but average, and so we need to understand the variation or range of demand placed on our process.

Here’s an example …

To build a demand profile take the following steps:

  1. Pick a product, product family, customer, customer-item pair, or business unit of interest.
  2. Determine an appropriate time unit – hourly, daily, monthly.
  3. Gather the true demand as best you can.  Be careful about using promise dates instead of requested dates, and be doubly cautious of schedules which are often smoothed, filtered, or otherwise manipulated.
  4. Create the graph or time series plot as above.
  5. Now calculate some simple descriptive statistics – range, minimum, maximum, standard deviation, etc.  In this example the average is 17 with a range of 49 and a standard deviation of 11.

What can we conclude?  Should we design our operations control around a demand rate of 17 a day?  Is the variation in demand something we can deal with?  How?

 

 

Tips for warehouse sizing


Tips for warehouse sizing:

  • Don’t get fooled by “averages”
  • Consider using statistical tools, such as standard deviation, and P.90 probability to analyze operational data – both in and out bound
  • Understand which system components can expand capacity by adding labor, and which can’t
  • Design expansion capability in from the start; SKU count almost always goes up, not down over time
  • Get executives to sign off on future sales projections that will serve as the basis of the design; if they won’t or can’t, then round up
  • Be very leery of unrealized plans to increase inventory turns; easier said then done
  • Consider ability to add overtime and additional shifts to expand initial system capacity
  • Recognize more companies regret having less capacity than those that think systems were over-specified
  • You can usually add labor to increase throughput in pick modules, but if a sorter is maxed out, there is not much you can usually do

ABC Analysis: how to

ABC Analysis is a common approach for prioritizing, classifying, or categorizing inventory management techniques. Classification is usually based on price or cost times usage or consumption volume. Typically ‘A Items’ are the highest dollar volume and represent 5 to 10% of the items and 50 to 70% of the total dollar volume. The ABC Analysis Principle is that A items are treated differently than C items.  A Items might be counted more frequently and have different planning parameters, higher inventory turns, higher safety stock and customer service levels.

  1. Make a list of part numbers
  2. Determine total quantity used over some period of time
  3. Obtaining the cost for each part
  4. Calculate usage $ value for each part by multiplying the quantity and the cost
  5. Sort the list from high to low $
  6. Calculate the total usage $ value for all items
  7. Calculate each item’s percent of total usage $ value
  8. Select percentage cut offs for each ABC group, for example:

 

Here’s an example …

Example of ABC Analysis

 

Next steps

Once you have classified your parts you can use this data to drive key materials management activities. For example, coordinating your perpetual inventory cycle counting program – you might routinely verify your Category A parts on a monthly basis but only review your category C parts twice a year.

You might use flow orders, kanban, or VMI for your C parts but require detailed negotiated purchase orders for your A parts.

In a warehouse you might want to be sure the A items are near the shipping dock and the C items are toward the back.

You might even want to take a close look at the C items and purge a few.

The main point is – one size doesn’t fit all parts, choose the materials management approach that best serves each inventory category.

Bullwhip Effect

The bullwhip effect is the result of uncertainty caused from distorted information flowing up and down the supply chain.  The bullwhip effect is caused by fluctuations in information supplied to firms further up the supply chain. Distorted information causes firms to forecast demand incorrectly.  Thereby, many unnecessary costs are put upon each of the firms along the supply chain.  Nearly all industries are affected!  Firms that experience large variations in demand are at risk.  Firms that depend on suppliers upstream or distributors and retailers downstream may be at risk.   Most firms are affected by the bullwhip effect.  The bullwhip effect used to be considered a normal phenomenon.  However, recently, many firms have been trying to focus on how to improve communication along the supply chain.  The bullwhip effect can inflict many unnecessary costs on business firms.  Inventory costs from stored inventory, problems with quality caused from rapid production, overtime expenses for increased employee labor, and increased units being shipped create costs far and beyond normal levels of production.  Customers can also lose faith in a firms ability to deliver products.  This is because firms are having trouble meeting demand.  Likewise, firms often must lengthen lead time for finished goods, which also may discourage customers, which in turn leads to lost sales.  In a worst case, incorrect forecasts may entice a company to adjust capacity which could be detrimental to the overall success of the company.  To reduce stocked product, retailers may offer sales promotions to customers.  If retailers fail to notify firms upstream in the supply chain, these firms may forecast increased sales as legitimate demand.  Thereby producing product that was not wanted by the customer in the first place.  Furthermore, salesforce incentives may entice selling products to firms to meet targets.  This may cause large inventories for the firm, or the firm may cancel the orders, which causes demand fluctuations in the supply chain.   Firms upstream in the supply chain may feel that the increased demand may be legitimate and increase production and inventory levels to produce more.  However, in reality, the product hardly moved and required a drop in price to be moved off of retailer’s shelves.  Each firm upstream in the supply chain will feel the whip effect.

Here’s the classic illustration from The Bullwhip Effect in Supply Chains by Hau L. Lee • V. Padmanabhan • Seungjin Whang, SLOAN MANAGEMENT REVIEW/SPRING 1997.

 

Supply Chain Strategy

 

 

Here’s a diagram of the effect of lead time and demand uncertainty on supply chain strategy.  Pull strategy works when lead times are short and high demand uncertainty makes building to forecast wasteful.  Dell is an example of this approach.  The number of feature combinations is high and customers are willing to wait a few days to get exactly what they want.  Push makes sense when lead times are long and demand is stable.   Canned soup is an example of where customers won’t wait for the fresh soup to be made and delivered.  Books were examples of the push strategy, but are now moving toward pull as print-on -demand and ebook readers proliferate.  When demand is known and lead times are short Continuous Replenishment or rate-based supply make sense.  Suppliers get point-of-sale data to release shipments on an agreed upon frequency to maintain inventory targets.  Throughout the continuous replenishment supply chain production and distribution operate on pull, and push at the retail outlets.  Positioning inventory strategically is complicated when lead times are long and demand is uncertain.  Strategic Inventory analysis can help sort out how much stock to carry, and where.  The point in the supply chain where risks can be pooled or demand aggregated is usually the place for some statistical safety stock.

Safety Stock Optimization

Many of you are looking for a correct, comprehensive safety-stock calculation. My company’s approach is based on more than 15 years of development and testing. As we have learned through extensive experience, optimized safety stock requires more than a formula. We use complex demand–data modeling and calculations. We provide a service, not a spreadsheet. You send us your data. We send you the results.

Our safety-stock model is correct and comprehensive, providing optimal safety stock levels for your service-level and financial-performance targets. We include all the factors that affect safety stock and service level, and apply the proper statistical techniques to them. We do not utilize the usual stockout-event-based metric, but the same quantity-based fill-rate criterion that most companies use to measure actual service-level performance during a month, quarter or year. Our calculations represent the right-skewed and sporadic patterns typical of real demand data. Our model also includes past-due demand and its disruptions, probability of past-due-demand cancellation, lead time, reorder quantity (MOQ, EOQ, etc.), package size and reorder-review frequency. Finally, our results provide a high degree of confidence, typically 95%, of consistently achieving your target service levels without costly expediting.

For more details, see our white papers at www.topdownleansystems.com/white.htm. Page 16 of the “Common Safety Stock Calculations” white paper has examples of our safety-stock analysis. Also, see how you do on our Safety Stock Quiz, at www.topdownleansystems.com/quiz.htm.

To demonstrate the power of our approach, we would be happy to calculate and analyze safety-stock levels for a sample of your inventory items at no charge. Send me data on up to 30 of your items, and we will send you results – safety stock quantity and safety stock days for each item. Our analysis also includes each item’s range of expected actual performance for fill rate; average reorder and on-order quantities; average quantity on-hand, days on-hand and inventory turnover; average daily demand and demand activity percentage.

We require this data for each item: Item identifier, target fill rate, reorder quantity (MOQ, EOQ, batch size, lot size, etc.) or reorder frequency, package size (order multiple), lead time, probability of past-due-demand cancellation, days in actual service-level measurement cycle, and as much historical daily time-series demand as possible (three years is best, two is better, one is good). We perform extensive pre-screening on your input data to identify potential issues and to avoid “garbage in, garbage out.”

Send me your contact information via www.topdownleansystems.com/contact.php. I’ll provide you with a file containing input-data examples. Of course, I’ll be happy to explain our model in more detail and to answer your questions, at your request.

David McPhetrige, TopDown Lean Systems

Things to do to improve warehouse productivity

warehouse 4Business may be slow now but before you know it you’ll be jammed again. Want to get more done with the folks you have?  Things to consider:

  1. Keep the lifts in good repair
  2. Batteries getting old?
  3. Stagger shift starts – replenish forward picking before the first wave
  4. Reslot often – move the A items closer to the dock
  5. Not enough space, then make more – get to 10% empty forwards and 20% empty reserves
  6. Qualify and prioritize the inbound freight – need the trailer now, or later?
  7. Qualify the product going into reserve
  8. Get the inbound current and under control before tackling pick, pack, ship
  9. Fix any and all inventory inaccuracy root causes
  10. Have fresh eyes look at the problem – select different supervisors or warehouse workers to look at other areas
  11. Eliminate touches – live load, don’t pick and stage
  12. Minimize travel – never travel empty – put one away, pick one to ship
  13. Right size the forwards, so inbound doesn’t need to go into reserve
  14. Align the picking method for each product with its order pattern
  15. Can the WMS round up order quantities to an easily picked unit or measure?
  16. Engage the troops
  17. Every DC worker makes thousands of decisions each day; understand and guide discretionary decision-making
  18. Solve the workforce’s boredom problem
  19. Most supervisors spend less than 5% of their time on motivating employees, double that and double productivity
  20. Inbound congestion means waste and extra touches
  21. Housekeeping
  22. Address the annoyances that demotivate
  23. Keep inbound under control and putaway as timely as possible
  24. Recalculate Safety Stock
  25. Update leadtimes
  26. Bust the inbound batch sizes
  27. Increase inbound visibility, smooth the spikes if you can
  28. Publish metrics for all to see and encourage friendly competition between zones, departments, facilities
  29. Create a ‘dog pound’ and move slow movers out of the way
  30. Study and fight outbound congestion
  31. Adjust the number of pick zones; fewer the better
  32. Synchronize order filling across all zones
  33. Keep current on replenishment
  34. Never run out of supplies (totes, pallets, carts, tape)
  35. Adjust the organization chart
  36. Constantly monitor outbound flow; rebalance pick, pack, and loading
  37. Reduce the number of job classifications by cross training and rotation
  38. Use inbound teams and eliminate staging areas: unload, receive and put away with one touch not two or three
  39. Brainstorm and then brainstorm some more
  40. Be careful what you measure
  41. If you are in a meltdown, get help
  42. Consider postal pick location address scheme; going down an aisle picking on left and right instead of down one side and coming back the other

 

Waterspiders as continuous improvement innovators

water_beetleThe term Waterspider or water beetle (mizusumashi in Japanese) comes from the behavior of the insect known in the States as a whirligig, an aquatic animal that skitters around on the top of a pond quickly changing direction as it goes.  For a lean enterprise the role of material handlers, expediters, and support staff changes. In the Toyota Production System this is the common name for a person assigned to support a production operation, so that others may focus exclusively on value-added work. The waterspider delivers parts to the other associates in the cell or on the line so that they don’t need to stop to replenish their work stations.

Unlike a ‘floater’, a waterspider is assigned specific tasks, such as replenishing raw material inventories (via milk run), common area clean-up, communicate status, maintain visual metrics, etc… Waterspider duties usually don’t include tasks which take them away from the production area, or detract from their specific, assigned duties (the waterspider is not the ‘5S’ person or a ‘fill in’). Think of the waterspider as the ‘race car pit crew’ for the production team, without which it would be impossible to win or even run the race.

Waterspiders quickly become experts in the withdrawal and production kanban system. They can ‘see’ more of the up and down stream flow in real time than most others, and because of this often making it possible to identify and eliminate errors. From recent experience the waterspiders often have a better grip on reality than their managers, planners, and engineers.

Non manufacturing examples abound in restaurants, hospitals, insurance claims processing; serving the folks that add the value isn’t just for manufacturing. In product and software development the role of the program manager is sometimes something like that of the waterspider, except bringing knowledge to the various development team members instead of parts.

Here are a few references:
Single piece flow at ConMed Linvatec
Inventory management in electronics manufacturing: The Move To Lean
Lean in the Oil Fields

Have any examples you’d like to share?