Building applications, especially custom ones, carries with it the burden of answering the question: Does this do what the customer wants?

With complicated systems with many interacting features and business rules, answering this question can be daunting. In fact, evaluating the answer can be daunting too, from the perspective of the customer. Having the sales guy check some boxes in a questionnaire, or watching a demo just doesn’t leave you with the assurance that the application with handle all the business requirements, from either perspective, the vendors or the customer. Everyone I have spoken to who has sold complex software, or who has participated in the purchasing process of software has expressed the same doubt. They are just not sure that the tool will be a good fit. As we all know, that doubt does not always prevent the purchase of the software, as each organization has its own level of risk tolerance, and trust in the vendor’s brand or reputation. Often these other considerations can outweigh the amorphous doubt that some folks might feel. How can one quantify that doubt? Frankly, it’s a quandary.
This thought got us at Profit Point thinking… Wouldn’t it be great if there was another way to evaluate the goodness of fit or an application, or the appropriateness of the parameter settings, to match the business needs of an organization. Would it be great if there was a way to eliminate (or greatly reduce) the doubt, and replace it with facts. Either a business rule is obeyed or it is not. Either a decision is made according to the requirements, or it is not. Let’s eliminate the doubt, we thought, and the world would be a better place. (well a little bit anyway).

There are many processes for testing an application as it is being developed, with writing test scripts, and evaluating the results. All these are based on testing little pieces of code, to ensure that each function or sub routine does what it should do in each case of input data. These processes work fine in our opinion, but only when the sub of function is able to be considered independently form the others. When the system has functions that interact heavily, then this approach doesn’t reduce the doubt that the functions may conflict or compete in a way that the whole system suffers. How then to evaluate the whole system? Could we treat the entire application as one black box, and evaluate the important business cases, and evaluate the results? This is exactly what we have done, with the effect of reducing the doubt to zero about the suitability of the application for a business.
With several of our clients we have worked out what seems to be a great process of testing a complex software solution for suitability to the business requirement. In this case, the detailed level function testing methods were not open to us, since the solution relied on a Linear Programming technique.
This process is really just an amplification of the standard testing process.

1. Define the test case, with the expected results
2. Construct the test data
3. Build or configure the application
4. Run the Test using the Test Data and Evaluate the results – Pass or Fail

This is the standard process for testing small functions, where the expected results are clear and easy to imagine. However, in some systems where there many interacting rules and conflicting priorities, it may not be simple to know what the expected results should be without the help of the tool’s structure to evaluate them. Such is the case with many of our application, with layer upon layer of business rules and competing priorities… The very reason for using an LP based approach makes testing more complex.
In the revised process, we have, for each new business requirement:

1.  Construct the test case with the test data
2. Build or configure the application
3. Set the expected results using the results of the first pass build
4. Re-factor the code and test until all test are passing

Profit Point’s Software Testing Process

In my next blog I will show you the simple excel based tools we use to facilitate the test evaluation.

In practice, the process works well, new versions of the application go into production without any surprises, and with full confidence of the application management team that all the business requirements are 100% met.

No doubt – no doubt a better process.

By Jim Piermarini

There is nothing fundamentally new in the area of Supply Chain management! Supply Chains have existed ever since some caveman somewhere decided to make specialty dinosaur clubs from a particular kind of really hard wood and very sharp rocks that he got from two other troglodytes over in the next valley. What has changed however is the nature and speed of the communication that occurs between the participants in a Supply Chain, and the ability of those actors in the process to keep and use past information for making decisions.
In my current job in Supply Chain consulting, I frequently work with Production Schedulers. These are folks who I have a lot of empathy with, since once upon a time, a long time ago I actually started my career as a production scheduler (and no I was not involved in the dinosaur club market!).  And I was recently thinking about this idea that what is new in the area of supply chain management is a result of the way we communicate today as compared to the past.
I hate to admit it, but “back in the day” when I was scheduling, computers were huge boxes controlled by a phalanx of support people in separate departments, if not different buildings. They were not the practical job-aids that they are today. And if you used the word internet, people would have thought you meant to say ‘hairnet’.  In building a schedule the basic tool was a board where I would physically place magnetic strips that were marked to indicate which products should be produced on particular pieces of equipment.
The communication that occurred was in the form of either paperwork carried in the interoffice mail, phone calls, or face-to-face conversation. So if a particular manufacturing process was running late, there would usually be a tremendous time lag before the scheduler would find out. Or if there had been some major snafu somewhere either in my own organization, or in any of the suppliers’ plants, I probably would not hear about it until I proactively asked. And of course from the customer’s point of view, they couldn’t simply change their order and email (what’s that?) me the new information. Obviously there was no real-time tracking of a truck’s location, and whether or not it was broken down or stuck in traffic on the N.J. Turnpike. And of course because there were no computers, I could never really be sure as to just how much product was available to ship to customers. Although the paperwork might say that there were 20 pallets on hand, the guys out in the warehouse might have lost track of where exactly all of those pallets might be.
What this all inevitably led to was a lot of extra inventory being carried at every stage of the Supply Chain. This was because you always wanted to try to have a little extra cushion built in to cover yourself for the unexpected. As a scheduler, I would catch a lot of grief (a nice way of saying you know what) from my boss if a production line was to shut down because they ran out of the intermediates or raw materials needed to keep running. Or if I committed to a customer order of 15 pallets on Wednesday, but the shipping department could only find 14 pallets, there was “hell to pay”.  I can remember literally climbing over the tops of shifting and unstable pallet stacks (OSHA would have had a field day), and shining a flashlight down the gaps looking for inventory that a “3 x 5” index card said was still in the warehouse somewhere.
And I don’t think the comparison of my experience vis-à-vis schedulers today, is any different than the comparison of those who did scheduling in 1900 versus me in 1974. In 1900, telephones were only just coming into widespread use, and typewriters were a comparatively new-fangled gadget. Ultimately, Trog (the son of the caveman who I mentioned earlier) was able to streamline his father’s Supply Chain tremendously when we was able to get the customers, as well as the wood and the rock guy, to start using that new concept called “writing” (of course which language to use could have been an issue).
The field of Supply Chain management will always be at the mercy of communication technology. Over time, Supply Chains will continue to morph into more efficient contributors to organizations’ bottom lines as the ability of humans to communicate with each other evolves.

Here’s an audio interview with Dr. Alan Kosansky on the “Future of Supply Chain Management”.

TRANSCRIPT:

Interviewer: What’s the future of supply chain management? Many companies have implemented ERP software solutions, but if you’re relying on well-traveled, standardized software to manage your supply chain, you could actually be eroding your competitive edge. Joining us now to explain why is Dr. Alan Kosansky, co-founder and President of Profit Point. Alan, welcome!

Now, Alan—ERP Software has definitely become commonplace as a solution in supply chain management—it’s certainly convenient, but is the software on its own enough?

Kosansky: ERP software plays a critical role in the enterprise. From its inception it has provided the backbone for accounting and financial functions. As it has extended into supply chain functions, it allows us to quantitatively manage the supply chain. All these systems have enabled significant efficiencies for companies over the past 20 years. And they have become commoditized. Leading companies are both leveraging what these ERP have to offer AND ALSO defining complementary supply chain processes that offer competitive advantage. For those supply chain processes for which being as good as the marketplace is enough, out of the box ERP and APS solutions are great. However, for those supply chain processes where your company believes they can create and maintain competitive advantage, using the solutions that the marketplace is using is not enough.

Interviewer: At Profit Point you believe that the future of supply chain management is in optimization based decision making – what is optimization based decision making?

Kosansky: Supply Chain profitability is based on the price you sell your goods minus the total delivered cost of making and getting those products to your customers. While this may seem like simple arithmetic, it is actually very difficult for companies to accurately predict profitability and then make supply chain planning decisions that maximize their profitability. Firstly, Computing the total delivered cost is difficult. Secondly, even those companies that are have a centralized way to view all this data typically have difficulty making the tradeoffs implicit in their supply chain costs: Inventory or customer service? Manufacturing, warehousing or transportation costs? Optimization based decision making allows supply chain planners to both see all the relevant data and make the tradeoffs that lead to maximum profitability.

Interviewer: … and how can optimization based decision making help ‘unlock’ a company’s competitive edge?

Kosansky: Companies that identify supply chain processes where they have developed some sort of competitive advantage need to embody those processes in enabling technology that support this better decision-making. Most often, this includes some form of optimization decision technology that quickly evaluates alternative scenarios and identifies those decisions that lead to maximum profitability. By combining the big data that is available today, with leading edge decision making technologies, leading companies are beating their competitors in every aspect of their operations, including the supply chain.

Interviewer: Well Alan this is great news – thanks for coming on and telling us about it! That was Dr. Alan Kosansky, President of Profit Point. For more information go to ProfitPT.com… that’s ProfitPT.com.

The most mature Supply Chain organization are engaging in a strategic process whereby they identify two categories of supply chain processes within their company: Category B are those processes where performing at the industry standard is sufficient. Typical processes in this category include accounts payable, accounts receivable, raw material sourcing, and freight contracting. Category A are those processes where they believe they have ideas and/or practices that give them competitive advantage. These processes might include S&OP, production scheduling, territory planning, or network design.

For supply chain processes for which performing at the industry standard is sufficient, standard technology solutions are sufficient as well. However for processes here there is an opportunity for competitive advantage, out of the box standard solutions will not do. Often mathematical optimization is a critical enabling technology for those supply chain processes that are a source of competitive advantage. Furthermore there is typically some level of customization required in order to uniquely capture the ideas and processes that embody the competitive advantage.

For example, consider the order fulfillment process in consumer electronics. Products in this industry become obsolete very quickly. For this reason, electronics suppliers to “big box” retailers like Best Buy and Walmart often operate in a back-order situation to reduce the probability of getting stuck with returns and obsolete stock. While standard solutions most commonly use a traditional first-in, first-out (FIFO) method to allocate inventory to orders, more advanced consumer electronics companies use more sophisticated approaches to determine how best to assign limited inventory to their customers. These approaches take into account customer priority, in-transit inventory and inventory already in the channel to determine algorithmically in what sequence and quantity to assign inventory to orders. Since they cannot simply fill all the orders of their biggest customers at the expense of the rest of their customer portfolio, they apply sophisticated business rules to balance the needs of all their customers while making sure their most important customer’s feel the pinch of inventory shortfalls the least. In this case, a custom solution for deploying in-transit inventory helps to “score” orders based on customer priority as well as on the inventory in the channel. This approach, which uses logic and algorithms well beyond the capabilities of standard ERP solutions, reduces the seller’s total supply chain costs and improves its performance scorecard relative to its most important customers.

Mature supply chain organizations typically identify about 80% of their process as “standard” and are able to use out of the box standard ERP and supply chain solutions here. In addition, they are identifying about 20% of their processes as being a source of competitive advantage and are developing and implementing solutions that capture that advantage.

If your organization is engaged in these strategic management practices, we commend you for your leadership in supply chain maturity. If not, let’s work and grow together.

For a more detailed discussion of this topic, read the Supply Chain Quarterly magazine article by Dr. Alan Kosansky and Ted Schaefer entitled Is standardized software eroding your competitive edge?

This quarter’s Supply Chain Quarterly magazine features an article by Dr. Alan Kosansky and Ted Schaefer entitled Is standardized software eroding your competitive edge?  The article addresses the pros and cons of standard enterprise software packages and discusses how generic applications may not accommodate the processes that leading company’s utilize to gain competitive advantage.

Manufacture and delivery of a company’s products usually consume a wide array of materials, either directly or indirectly, ranging from rare commodities like titanium or zinc, to the most basic, such as water. Given the explosive growth of world population in recent history, and the resulting increases in consumption of food and other products, and the finite nature of raw materials, the sustainability of the supply chain over time is a growing planning concern for many companies.  Water is often a key focus in their planning, whether it is the main ingredient in their product, as it is in the beverage industry, or a major component, as it is for power generation, paper production, mining and many other industries.

One way to measure the water impact of companies (or countries, or production of industrial or agricultural products, such as textiles, rice or beef) is through the calculation of a “water footprint”, which can help identify what water is used (both directly and indirectly), where it comes from, and the relative efficiency of its use.  This concept is discussed in detail on the website www.waterfootprint.org  which has a wide array of statistics, as well as an interactive water footprint calculator and the option to download extensive research materials.  According to the website 92% of total water consumption in the world is associated with agricultural use.  However, since agricultural products are raw materials in many corporate supply chains, and are shipped from one location to another around the world, nations and companies effectively consume water from around the world.  The figure below shows major international water consumption flows, taking into account such factors as goods consuming water in production in one part of the world are shipped to a consumer in another area.

Source:Mekonnen and Hoekstra (2011)

Why should a company be concerned about their water consumption?  There are several risks that all companies face, to varying degrees, as global water consumption increases, including

• Physical supply risk: will fresh water always be available in the required quantities for your operations?
• Corporate image risk: your corporate image will likely take a hit if you are called out as a “polluter” or “water waster”
• Governmental interference risk: governmental bodies are becoming increasingly interested in water consumption, and can impose regulations that can be difficult to deal with
• Profit risk: all of the above risks can translate to a deterioration of your bottom line.

But with risk comes opportunity – planning for your water consumption, and footprint, as part of your supply chain analysis, and acting in response, can keep you ahead of the curve!

It wasn’t very long ago that the United States was still a major location for the manufacture of electronics. Apple Inc. used to boast that its products were all made in America. There were a number of large plants (and many were in fact foreign owned) making everything from televisions to mass-market audio devices.

But in the past few years that has changed dramatically. Today, almost all of the millions upon millions of Apple Inc. devices are manufactured oversees. Or look at the Amazon Kindle. Despite the fact that the key innovation that underlies the success of the Kindle – the electronic ink which is produced in Massachusetts – all the remaining components in this product are manufactured in Asia. Over dinner in February 2011, President Obama is rumored to have asked Steve Jobs of Apple Inc. why couldn’t the company’s products be “made in the U.S.A.”. Jobs replied “those jobs aren’t coming back” according to another dinner guest.

No company’s supply chain exists in a vacuum; they are like living organisms that exist and depend on their environment and surroundings. So what Jobs was saying was that the supporting economies, societies and infrastructure in which Apple’s supply chain exists have moved to Asia and it is an extremely difficult task to uproot it and relocate it to the U.S.

Executives at various mass-market electronics manufacturers tell glowing stories of the flexibility and responsiveness of their Asian suppliers. There’s a story where Apple made a late-stage design change to the frames around the iPhone screen. The Chinese manufacturer roused 8000 workers in the company’s dormitories at midnight, and within half an hour they began a 12-hour shift. Four days later, the factory was turning out 10,000 iPhones per day.

And the reason why these supply chains have taken root in Asia is not just a result of relatively cheap labor. In an article in the New York Times, Timothy D. Cook of Apple explained that factories in Asia “can scale up and down” at breathtaking speeds. They have the mid-level engineering talent and other skilled and un-skilled personnel resources to be able to rapidly adjust to their customer’s requirements. And in addition, the supporting 2nd-tier businesses that supply the electronics manufacturers are located nearby and are themselves extremely flexible and responsive. Thus a full and complete ecosystem has grown and flourished in East Asia for manufacturing mass-market electronics which is not easily or quickly replicated elsewhere in the world.

The experience of the electronics industry should be a cautionary tale for the U.S. and other advanced economies. Supply Chains are networks of interdependent actors. And in the case of manufacturing enterprises, their physical location has an impact on their ability to perform efficiently. This nation should have a debate as to what are the critical industries that we want to keep rooted in this country and then develop policies and infrastructure that will foster the growth and long-run health of the businesses that are involved in these areas of the economy.

There is nothing like a bit of vacation to help with perspective.

Recently, I read about the San Diego Big Boom fireworks fiasco — when an elaborate Fourth of July fireworks display was spectacularly ruined after all 7,000 fireworks went off at the same time. If you haven’t seen the video, here is a link.

And I was reading an article in the local newspaper on the recent news on the Higgs: Getting from Cape Cod to Higgs boson read it here:

And I was thinking about how hard it is to know something, really know it. The data collected at CERN when they smash those particle streams together must look a lot like the first video. A ton of activity, all in a short time, and a bunch of noise in that Big Data. Imagine having to look at the fireworks video and then determine the list of all the individual type of fireworks that went up… I guess that is similar to what the folks at CERN have to do to find the single firecracker that is the Higgs boson.

Sometimes we are faced with seemingly overwhelming tasks of finding that needle in the haystack.

In our business, we help companies look among potentially many millions of choices to find the best way of operating their supply chains. Yeah, I know it is not the Higgs boson. But it could be a way to recover from a devastating earthquake and tsunami that disrupted operations literally overnight. It could be the way to restore profitability to an ailing business in a contracting economy. It could be a way to reduce the greenhouse footprint by eliminating unneeded transportation, or decrease water consumption in dry areas. It could be a way to expand in the best way to use assets and capital in the long term. It could be to reduce waste by stocking what the customers want.

These ways of running the business, of running the supply chain, that make a real difference, are made possible by the vast amounts of data being collected by ERP systems all over the world, every day. Big Data like the ‘point-of’sale’ info on each unit that is sold from a retailer. Big Data like actual transportation costs to move a unit from LA to Boston, or from Shanghai to LA. Big Data like the price elasticity of a product, or the number of products that can be in a certain warehouse. These data and many many other data points are being collected every day and can be utilized to improve the operation of the business in nearly real time. In our experience, much of the potential of this vast collection of data is going to waste. The vastness of the Big Data can itself appear to be overwhelming. Too many fireworks at once.

Having the data is only part of the solution. Businesses are adopting systems to organize that data and make it available to their business users in data warehouses and other data cubes. Business users are learning to devour that data with great visualization tools like Tableau and pivot tables. They are looking for the trends or anomalies that will allow them to learn something about their operations. And some businesses adopting more specialized tools to leverage that data into an automated way of looking deeper into the data. Optimization tools like our Profit Network, Profit Planner, or Profit Scheduler can process vast quantities of data to find the best way of configuring or operating the supply chain.
So, while it is not the Higgs boson that we help people find, businesses do rely on us to make sense of a big bang of data and hopefully see some fireworks along the way.

Isn’t that one of our main objectives in life, whether the setting is business, participation in sports, your personal life?

I see part of our role at Profit Point as helping our clients to achieve their potentials. We do this by applying mathematical techniques to find good solutions to the problems that business leaders face. Many of our clients call upon us when their business is going through a time of transition, particularly when there is a merger of organizations.

Analyzing the potential for facility rationalization is one of the standard uses of our Profit Network infrastructure planning software. We, and clients, have used this software to decide how many plants, production lines and warehouses they need to best serve their customers in many different types of situations.

But mergers present opportunities to organizations further down the supply chain as well, of course. Many companies use vehicles to deliver product to customers on a regular basis, and when there is a merger (and at other times) well-run businesses are looking for ways to ensure that these types of activities are carried out efficiently.

Our Profit Vehicle Planner (PVP) software can help in planning for a merger at that next level down – for instance, when you have two organizations serving customers in a metro area, how do you combine them together?

The diagrams below give you an idea of the situation a company might face. They have operations in various parts of the country, serving hundreds of customers in each area. Their Southern California customers might be spread as in the pattern in the diagram below on the left.

To serve these customers they currently have five route territories, covering the customer deliveries, as is shown in the diagram on the right.

Now they plan to merge with a smaller competitor in the same type of business. The acquired company has customers in southern California with a similar spread across the geography, divided into two territories, as is shown in the diagrams below.

PVP will allow the analyst to look at all of the customers together,

and in this case, when the territory planning algorithm runs, it finds that deliveries can be made in six more-compact route territories, covering all customers. Separately the two companies had seven territories – and merged they have the potential to serve them with six – thus saving a truck and various associated expenses.  The merged solution is shown below.

Implementing this merged solution can help the company better achieve its potential – for profits.

Uncovering the Value Hiding Behind Environmental Improvement Investments

Supply Chain optimization is a topic of increasing interest today, whether the main intention is to maximize the efficiency of one’s global supply chain system or to pro-actively make it greener. There are many changes that can be made to improve the performance of a supply chain, ranging from where materials are purchased, the types of materials purchased, how those materials get to you, how your products are distributed, and many more. An additional question on the mind of some decision makers is: Can I minimize my environmental footprint and improve my profits at the same time?

Many changes you make to your supply chain could either intentionally – or unintentionally – make it greener, so effectively reducing the carbon footprint of the product or material at the point that it arrives at your receiving bay. Under the right circumstances, if the reduced carbon footprint results from a conscious decision you make and involves a change from ‘the way things were’, then there might be an opportunity to capture some financial value from that decision in the form of Greenhouse Gas (GHG) emission credits, even when these emission reductions occur at a facility other than yours (Scope 3 emissions under the Greenhouse Gas Protocol).

As an example, let’s consider the possible implications of changes in the transportation component of the footprint and decisions that might allow for the creation of additional value in the form of GHG emission credits. In simple terms, credits might be earned if overall fuel usage is reduced by making changes to the trucks or their operation, such as the type of lubricant, wheel width, idling elimination (where it is not mandated), minimizing empty trips, switching from trucks to rail or water transport, using only trucks with pre-defined retrofit packages, using only hybrid trucks for local transportation and insisting on ocean going vessels having certain fuel economy improvement strategies installed. These are just some of the ways fuel can be saved. If, as a result of your decisions or choices made, the total amount of fuel and emissions is reduced, then valuable emission credits could be earned. It is worth noting that capturing those credits is dependent on following mandated requirements and gaining approval for the project.)

Global Market for GHG Credits

If your corporate environmental strategy requires that you retain ownership of these reductions, then you keep the credits created and the value of those credits should be placed on the balance sheet as a Capital Asset. Alternatively, if you are able, the credits can be sold on the open market and the cash realized and placed on the balance sheet. Either way, shareholders will not only get the ‘feel good’ benefit of the environmental improvement, but also the financial benefit from improvement to the balance sheet. If preferred, the credits can be sold to directly offset the purchase price of the material involved, effectively reducing that price and so increasing the margin on the sales price of the end-product and again improving the bottom line. If capital investment is required as part of the supply chain optimization, the credit value can also be a way to shorten the payback period and improve the ROI, or to allow an optimization to occur

So, when you consider improving your environmental impact or optimizing your supply chain, consider the possibility that there might be additional value to unlock if you include both environmental and traditional business variables in your supply chain improvement efforts.

Written by: Peter Chant, President, The FReMCo Corporation Inc.

I was sitting on the plane the other day and chatting with the guy in the next seat when I asked him why he happened to be traveling.  He was returning home from an SAP ERP software implementation training course.  When I followed up and asked him how it was going, I got the predictable eye roll and sigh before he said, “It was going OK.”  There are two things that were sad here.  First, the implementation was only “going OK” and second, that I had heard this same type of response from so many different people implementing big ERP that I was expecting his response before he made it.

So, why is it so predictable that the implementations of big ERP systems struggle?  I propose that one of the main reasons is that the implementation doesn’t focus enough on the operational decision-making that drives the company’s performance.

A high-level project history that I’ve heard from too many clients looks something like this:

1. Blueprinting with wide participation from across the enterprise
2. Implementation delays
   1. Data integrity is found to be an issue – more resources are focused here
   2. Transaction flow is found to be more complex than originally thought – more resources are focused here
   3. Project management notices the burn rate from both internal and external resources assigned to the project
3. De-scoping of the project from the original blueprinting
   1. Reports are delayed
   2. Operational functionality is delayed
4. Testing of transactional flows
5. Go-live involves operational people at all levels frustrated because they can’t do their jobs

What is promised during blueprinting

What is delivered at go-live

Unfortunately, the de-scoping phase seems to hit some of the key decision-makers in the supply chain, like plant schedulers, supply and demand planners, warehouse managers, dispatchers, buyers, etc. particularly hard, and it manifests in the chaos after go-live.  These are the people that make the daily bread and butter decisions that drive the company’s performance, but they don’t have the information they need to make the decisions that they must make because of the de-scoping and the focus on transaction flow.  (It’s ironic that the original sale of these big ERP systems are made at the executive level as a way to better monitor the enterprise’s performance and produce information that will enable better decision-making.)

What then, would be a better way to implement an ERP system?  From my perspective, it’s all about decision-making.  Thus, the entire implementation plan should be developed around the decisions that need to be made at each level in the enterprise.  From blueprinting through the go-live testing plan, the question should be, “Does the user have the information in the form required and the tools (both from the new ERP system and external tools that will still work properly when the new ERP system goes live) to make the necessary decision in a timely manner?”  Focusing on this question will drive user access, data accuracy, transaction flow, and all other elements of the configuration and implementation.  Why? Because the ERP system is supposed to be an enabler and the only reasons to enter data into the system or to get data out is either to make a decision or as the result of a decision.

Perhaps with that sort of a focus there will be a time when I’ll hear an implementation team member rave about how much easier it will be for decision-makers throughout the enterprise once the new system goes live.  I can only hope.

Supply chain managers are tasked with making the critical decisions to improve supply chain operations by taking costs out of the system while improving customer service and profitability. The best managers rely not only on their experience but also on data-based decision-making. Making the best decision to minimize supply chain costs and maximize profit requires accuracy in your data.

However, don’t confuse accuracy with precision. Too many decision makers throw out the baby with the bath water when they shun data based decisions because the data does not precisely reflect the detail of their operations. Wise supply chain managers and analysts understand that for many decisions, aggregated and/or averaged data can accurately reflect the cost/benefit tradeoff of critical decision and point in the direction of near-optimal decisions. For example, modeling manufacturing capability and capacity at the product family level is often accurate enough to make the right supply and demand balancing decisions. Another example: even when optimizing detailed scheduling operations within a single plant it is important to find the right level of detail to model so as to balance accurately capturing the realities of the manufacturing operations and being able to evaluate a large number of production sequencing options in order to find the optimal schedule.

Finding the right level of data detail to inform accurate decision tradeoffs while searching for the optimal decisions is an art as much as a science. It requires experience and expertise in both supply chain operations and optimization modeling.

This is my passion, so let me know what experiences you’ve had, and how you have found that balance to achieve optimal performance in your supply chain operations.

Alan Kosansky
akosansky@profitpt.com
610-645-5557

I work in the area of Supply Chain management and optimization. In my world, the organizations I’m usually dealing with are relatively mature with pre-existing supply chains. Recently though, I’ve been considering the challenges that face a single entrepreneur who is trying to start a small business and must create a whole new supply chain from scratch. Just think about all of the information that this start-up must gather and digest before any decisions are made. Then based on this data the entrepreneur makes a range of decisions; any one of which could sink the entire enterprise. When you realize the amount of work and skill that goes into this process, you can’t help but admire the individual who embarks on such a journey.
What prompted these musings was that my nephew has set up a business that will import various teas and dried fruit from East Africa into the U.S (his company is Mavuno LLC, the brand is Mavuno Harvest). He spent some time in Kenya in the Peace Corps, so he’s familiar with the language and he does have contacts there. His first challenge was to find an African source who could meet U.S.F.D.A manufacturing standards. And a related issue was understanding U.S. labeling requirements (you know the ‘x% carbohydrates’ and ‘y% calories from fat’ kind of stuff), then finding a lab that can do this work, and getting samples of his product to the lab.
Next there was designing the packaging, and finding a printer who could supply the packaging to the manufacturer. There were no printers in Kenya who were interested in his small order (10,000 bags), so he went with one in India. And he used a website to design the packaging, who in turn put him in touch with a company in Croatia to actually make the stuff. And let’s not forget lining up some financing because none of these folks work for free. Once he had his first batch of product made, he had to arrange for it to be shipped back to the U.S. in one of those 20-ft shipping containers.
While this first load of product was making its way across the ocean and through the various customs offices, my nephew high-tailed it back stateside and started knocking on doors. He previously had some discussions in America with certain merchants, so he did have some sense that people would be interested in the product. His selling points are that it is “Ethically sourced” (which is code for the fact that the workers are fairly treated) from “Sustainable Agriculture”. It would be sold in places like coops and relatively high-end markets.
But now his next problem is how to deal with success; what if these products sell! Now he has to go back and create a functioning, dependable, repeatable Supply Chain. He would have to go to that printer and manufacturer in Kenya and get them to make more; maybe get a few openings in their schedules to make regular production runs . Once that happens, then my nephew would need to set up a regular mechanism for shipping the product. And when product begins arriving in the U.S., where does he store it? Maybe just drop the shipping container in the back of my brother’s house?
Obviously there are entrepreneurs out there, who are wrestling with these kinds of decisions every day. They’re building Supply Chains from the ground up and the long-term success of their enterprises depends on the decisions they make at the initial stages of the process. By luck and by design, they are putting the foundations of their Supply Chains in place, and it is these first steps which will dictate the future. I truly admire their initiative, and wish them the best of luck.

Change is hard.

Collapsed Souffle

In the businesses that I help, change comes for several reasons. It may be thrust upon the business from the outside, a change in the competitive landscape for instance, or a new regulation.   It may come from some innovative source within the company, looking for cost savings to increase profitability of productivity, or a new process or product with increased productivity. Change can come from the top down, or from the bottom up. Change can come in a directed way, as part of a larger program, or organically as part of a larger cultural shift.  Change can come that makes your work easier, or harder, and may even eliminate a portion (or all) of the job that you were doing. Change can come to increase the bottom line or the top line. But primarily change comes to continue the adaptation of the company to the business environment.  Change is the response to the Darwinian selector for businesses.  Adapt or decline. Change is necessary.  It is clear to me from my experience that businesses need to change to stay relevant.

This may seem trite or trivial, but accepting that change is not only inevitable, but that it is good, is the shift in attitude that separates the best companies (and best employees) from the others.

So, you say, I see the need to change, it is not the change itself that is so difficult, but rather the way that it is inflicted upon us that makes it hard.  So, why does it have to be so hard?  Good question.

Effective managers know that change is necessary but hard. They are wary of making changes, and rightly so.  Most change projects fail. People generally just don’t like it.  Netflix is a great example.  Recently, Netflix separated their streaming movie service from their DVD rental business. After what I am sure must have been careful planning, they announced the change, and formed Quikster, the DVD rental site, and the response from the customer base was awful. As you likely know, Netflix, faced with the terrible reception from their customer base and stockholders, reversed their decision to separate streaming from DVDs. What was likely planned as a very important change, failed dead. Dead, dead, dead. Change can be risky too.

If change is necessary, but hard and risky… how can you tame this unruly beast?

The secret of change is that it relies on three things: People, Process, and Technology. I name them in the order in which they are important.

People are the most important agents relative to change, since they are the one who decide on the success or failure of the change. People decided that the Netflix change was dead. People decide all the time about whether to adopt change. And people can be capricious and fickle. People are sensitive to the delivery of the change.  They peer into the future to try to understand the affect it will have on them, and if they do not like what they see…  It is the real people in the organization who have to live with the change, who have to make it work, and learn the new, and unlearn the old. It is likely the very same people who have proudly constructed the current situation that will have to let go of their ‘old’ way of doing things to adopt to the new. Barriers to change exist in many directions in the minds of people.  I know this to be true… in making change happen, if you are not sensitive to the people who you are asking to change, and address their fears and concerns, the change will never be accepted.  If you do not give them a clear sense of the future state and where they will be in it, and why it is a better place, they will resist the change and have a very high likely hood of stopping the change, either openly, or more likely passively and quietly, and you may never know why the fabulously planned for change project failed.

Process is the next aspect of a change project that matters.  A better business process is what drives costs down. Avoiding duplication of efforts, and removing extra steps. Looking at alternatives in a ‘what-if’ manner, in order to make better decisions, these are what make businesses smarter, faster, better.  A better business process is like getting a better recipe for the kitchen. Yet, no matter how good a recipe; it still relies on the chef to execute it and the ovens to perform properly. Every business is looking for better business processes, just as every Chef is looking for new recipes.   But putting an expert soufflé recipe, where the soufflé riser higher, in the hands of an inexperienced Chef does not always yield a better soufflé.  People really do matter more than the process.

Technology is the last aspect of the three that effect change. Better technology enables better processes. A better oven does not make a Chef better.  The Chef gets better when they learn to use the new oven in better ways, when they change the way they make the soufflé, since the oven can do it.  A better oven does not do it by itself.  An oven is just an oven. In the same way, better technology is still just technology.  It by itself changes nothing.  New processes can be built that use it, and people can be encouraged to use it in the new process.  Technology changes are the least difficult to implement, and it is likely due to this fact that they are often fixed upon as the simple answer to what are complex business problems requiring a comprehensive approach to changing the business via it people, process, and technology.

Nice Souffle

Change is necessary, but hard and risky. Without change businesses will miss opportunities to adapt to the unforgiving business world, and decline. However, change can be tamed if the attitude towards it is changed to be considered a good thing, and is addressed with a focus on people, process and technology, in that order.  Done right, you can implement the change that will increase the bottom line and avoid a collapse of your soufflé.

Rich Guy

The rise of zombies in pop culture has given credence to the idea that a zombie apocalypse could happen. In a CFO zombie scenario, CFO’s would take over entire companies, roaming the halls eating anything living that got in their way. They would target the brains of supply chain managers and operations people. The proliferation of this idea has led many business people to wonder “How do I avoid a CFO zombie apocalypse?”

Supply chain managers are seeking and developing new and improved ways to exploit the volumes of data available from their ERP systems. They are choosing advanced analytics technologies to understand and design efficient sustainable supply chains. These advanced analytics technologies rely on the use of optimization technology. I am using the mathematical concept of “optimization” as opposed to non-mathematical process of making something better.

Mathematical optimization technology is at the heart of more than a few supply chain software applications. These applications “optimize” some process or decision. Optimization-base programs, for example, those frequently found in strategic supply chain network planning, factory scheduling, sales and operations planning and transportation logistics use well-known mathematical techniques such as linear programming to scientifically determine the “best” result. That “best solution” is usually defined as minimizing or maximizing a single, specific variable, such as cost or profit. However, in many cases the best solution must account for a number of variables or constraints. Advanced analytics technologies can improve a company’s bottom line – and it can improve revenue, too! CFO’s like this.

Advanced analytics technologies provide easy-to-use, optimization-based decision support solutions to solve complex supply chain and production problems.  And, these solutions can help companies quickly determine how to most effectively use limited resources and exploit opportunities.

So, from my perspective, there are seven practical reasons to embrace advanced analytics technologies:

1. Your company saves money, increases profits.
2. You get to use all your ERP system’s data.
3. It’s straightforward and uncomplicated.
4. You have the tools to discover great ideas and make better decisions.
5. At the end of the day, you know the total cost of those decisions.
6. You have a roadmap to make changes.
7. You avoid the CFO zombie apocalypse

Imagine that you’re running a business that offers a wide variety of products ranging from security, to bus and ferry transportation, to public toilets, to refreshments. You have somewhere between 45,000 and 65,000 customers, who all arrive at your various locations en mass, requiring service over a very short time span. To further complicate your situation, you don’t get the opportunity to train many of your ‘employees’ on what they should or shouldn’t do; instead the best you can do is rely on their common sense and good intentions. Actually, since many are volunteers, you can’t necessarily be sure as to the actual number of people who will come to work for you. This is the business that the New York Road Runners Club will be in on November 6 during the running of ING New York City Marathon. And the lessons that the race organizers have learned for managing the event are of universal value and applicability to many logistics and supply chain organizations.

Consider food and water. There are about 23 locations along the route (about 1 per mile starting at mile 3), where mostly volunteer staff is responsible for “stocking” cups of water, Gatoraide, and PowerGel (only available at mile 18). In the past, certain of the less desirable stations have been undermanned, while at others as the day drags along some of the volunteers simply up and leave; “I’m a volunteer, so fire me”. And of course, once the race is over, the staff that remains need to perform the thankless and definitely unglamorous job of cleaning up (at least to some extent) the litter that has been created.

And then there is security. The race attracts a number of world-class runners and security precautions need to taken to make sure that fans and paparazzi do not become too intrusive. To insure that these individuals return year after year, the organizers make sure that these athletes are able to start the race at the head of the pack so that they don’t have to dodge other slower amateurs. In addition, a large number of people typically run the race (and utilize the services) without having officially registered and paid any money. This year, there were about 140,000 applications for the 62,000 slots in the race. This means that about 78,000 people were rejected, and it is estimated that about 15% of these will run the race anyway. The Road Runners Club does not try to prevent any of these unofficial entrants due to the chaotic nature of the race start. However, security is in place at the finish, when the runners are strung-out, so that these “bandits” as they’re called, do not receive any of the available memorabilia.

Finally there’s transportation. Since the race is a “point to point” one, where the finish is at a different location than the start, runners are provided transportation to the beginning of the race. About 20,000 participants will take 522 chartered buses to the starting line. And they’ll all need this service in the space of about 2.5 hours. Since there is only 1 available bridge, if one of the buses were to breakdown at a critical location, a major traffic jam could result delaying many more runners than just those on the one disabled vehicle. Alternatively, approximately 21,100 entrants are expected to use the Staten Island ferry system. Between 5:30am and 8:30am, there will be a ferry leaving Manhattan every 15 minutes. Although the largest of these can accommodate 6,500, the organizers will try to limit each individual trip to half of the ship’s capacity. Again, this is an attempt to make sure that in the event of a breakdown, the number of runners affected (either delayed in Manhattan or stuck on a boat in the middle of the harbor) will be minimized. In fact, one of the ferries did experience mechanical problems in 2010. And when the runners disembark the ferry, a fleet of 70 buses will shuttle them the 2 or 3 miles to the starting line: 10 buses will load simultaneously at the St. George Terminal for the short trip. Last year the average load time for each group of 10 was 4 minutes, 22 seconds.

The New York Road Runners Club typically receives wide praise on blogs and other feedback forums from participants, for how well the Marathon is organized and run. In light of the unique circumstances of such an event, they have learned how to run their logistics operation like clockwork, always anticipating and planning for the worst that might happen. They’ve figured out how to manage a supply chain that is spread over a wide area and with a “work force” that presents its own unique challenges. Many businesses would do well to study their methods and take a page from the same playbook.

I recently saw this short post on a supply chain-oriented LinkedIn group:
I am responsible for 17 warehouses around the GCC. I want to create 4 major “Hubs” instead. Which would you choose and why, centralized or decentralized warehousing?
This post elicited a variety of responses. Suggestions ranged from looking at the details
Is it a circle route covering all four hubs or direct to and from specific hubs? Because backhaul opportunities can impact the overall costs within your network.
to broader advice, such as
Success of this depends on demand of product category and lead time importance.
When I am faced with a supply chain network design problem, like the one implied here, my first step is to develop a clear definition of the question to be answered – not to hastily jump to a solution. Along with defining the question, you need to determine
• the objective (minimizing cost, maximizing profit, or something else?),
• the options available (the current and potential locations for warehouses in the area, in the context of the post above),
• the constraints that are of importance (service requirements, warehouse capacity, transportation lanes),
• the time horizon, and related criteria.
Then, in the case of a complex supply chain, I find that building a model and using it to evaluate options can give insight regarding both the quantities and the qualities associated with a given solution. Modeling requires effort – you need skills, as well as various data, such as the sales forecast, warehouse location options, transportation and duty costs, etc. – but with the help of a model, and good quality information to populate it, you are able to estimate the implications of different supply chain options, and know whether operating with four hubs is more cost effective than five, or three; where should the hubs best be located; or whether you should have hubs at all. A model can help you make evaluate a variety of trade-offs so that you arrive at a better-informed, and more profitable, decision on how to proceed.

Many commercial enterprises and public agencies operate telephone call centers to provide effective and timely service for customers. By employing nearly 5% of the national workforce, call centers arguably define the “new factory floor” in an increasingly service-based economy. They are fascinating socio-technological systems which are exceptionally well-suited for the application of mathematical modeling and optimization methods.

A typical call center utilizes a computerized call handling system which can archive detailed historical information on call volume, call handling time, and other relevant attributes. This data can be analyzed and aggregated (with appropriate accounting for probabilistic variation) to generate a profile of staffing requirements across future time intervals. In theory, service agents can be optimally scheduled to closely accommodate this profile, resulting in high service levels, low customer abandonment, and efficient agent utilization. In actual practice, however, such performance represents the exception rather than the rule. Most call centers, even well-run ones, do not simultaneously achieve high levels of service quality and operational efficiency [1].

One important reason for the performance gap between theory and practice is lack of sophistication and flexibility in the standard software systems available for call center management. For example, standard systems invariably base interval staffing requirements on the classic “Erlang C” model, which is known to produce distorted results because it does not consider pertinent factors such as customer impatience [2]. Additionally, if the software has any capability for schedule “optimization,” the underlying algorithm is usually a greedy heuristic which sequentially adds agent shifts without due consideration of the complex interactions between them. Beyond these technical limitations, standard systems offer minimal capability to experiment with different shift types and customize the solution strategy.

Profit Point can provide the expertise and custom tools necessary to properly model your unique call center environment and achieve optimal performance. By applying recently-refined mathematics, interval staffing requirements can be accurately determined and optimal shift distributions can be precisely derived [3]. Efficiency improvements exceeding 10% are typical, coincident with improvement in service level performance. Many additional operational factors, such as on-line chat activity and agent specialization, can also be addressed. There is no better time than now for you to reap the rewards of optimizing your organization’s call center operations.

References

[1] Noah Gans, Ger Koole, and Avishai Mandelbaum, “Telephone Call Centers: Tutorial, Review, and Research Prospects,” Manufacturing and Service Management 5, 79–141 (2003).

[2] Lawrence D. Brown, et al., “Statistical Analysis of a Telephone Call Center: A Queueing-Science Perspective,” Journal of the American Statistical Association 100, 36–50 (2005).

[3] Dennis C. Dietz, “Practical Scheduling for Call Center Operations,” Omega 39, 550–557 (2011).

Heuristics and Optimization

 You might think the title should be ‘Heuristics or Optimization’, implying a choice.  But often the two approaches work well together with heuristics speeding an optimization process.  The Wikipedia definition of heuristic calls it an experienced-based technique for problem solving, learning, and discovery.  Wikipedia also mentions using heuristics to find a good enough solution and describes them as ‘strategies using readily accessible, though loosely applicable, information to control problem solving.’

 Those descriptions do not emphasize another aspect of heuristics – there is generally an underlying concept that informs the heuristic.  There is a good reason why we think it will work well in the majority of cases.  For example, an angle sweep heuristic is often used when designing routes for pickup and delivery from a central hub.  Those routes are candidates for selection in a formal optimization.  The designed routes look a lot like the petals of a daisy. 

 The heuristic starts out by heading north and picking locations close to that direction on the way out and back.  How far out the route goes is a property of vehicle capacity or time limitations.  The next route to be generated starts out slightly east of north and follows the same limitations and usually overlaps many of the locations on the first route.  Once the entire compass has been swept, the best set of routes to cover all locations is selected by an optimization.  In the example the heuristic becomes a front end for the optimization. 

 Another example comes from a driver scheduling problem.  Suppose a set of drivers must pick up some commodity from a set of locations for processing at a central plant.  Each trip in this example is an out-and-back because of the nature of the commodity, i.e. only one location can be visited.  Drivers pick up multiple loads in a day, and each location requires multiple visits.  The pickup times are fixed because of other problem features.  One approach is to simply allow all combination of driver-load-location pairings and let an optimizer grind away. 

 But there are other desirable features of the solution:  equalizing number of loads among drivers, and keeping driver dead time between loads to a minimum.  Specifying all the driver loads by some simple heuristic, e.g. send a driver out for the next load as soon as possible, usually ends up with some loads that cannot be covered.  A totally greedy approach fails.

 An approach that seems to work well in this case is to consider some drivers for early loads and some for late loads.  Work from the front of the early loads assigning each of the early drivers the first two loads they can feasibly complete.  Then work from the end of the late loads assigning the last two loads of the late drivers as the last two loads they can feasibly complete. 

 The loads in the middle and the drivers that are not considered early or late are handled by the optimization.  Notice that the heuristic does well on the driver gaps and guarantees that most drivers automatically get two loads, which is a good base in this application.  It also serves to speed the optimization by reducing the pairings to be searched while preserving enough flexibility to get a solution. 

 Furthermore, the heuristic is flexible in that one can choose how many drivers to consider early or late and how many of their loads to nail down heuristically.  Most importantly the heuristic gets better solutions than the optimization finds in any reasonable time.  So while the optimization must have that best solution out there, it will not find it in the time frame the scheduler has to work within.

 My experience is that flexibility is one of the key properties in any good heuristic.  Adaptability to new situations is also a feature of good heuristics.  One final example illustrating adaptability is based on an algorithm called ‘Toyoda’s Algorithm’.  I have applied this particular idea to a number of situations.

 In this example it is applied to sequencing the unloading of containers at a port.  Each container holds a selection of parts which have to be processed by various work centers prior to shipment out of the port.  A manifest shows the selection of the parts and it is known how much work is associated with a given part at a work center.  Not every work center can process every part type.  The objective is to get all the work centers to end their day at the same time and to keep all the work centers busy throughout the day.

 The approach is easy to understand in two dimensions.  The X and Y axis represent the available work time of two work centers, e.g. eight hours.  The arrows represent the amount of work delivered to a work center by a given container.  The dashed line is the ‘ideal path’ — equal amount of work at each work center throughout the day.

 The heuristic simply needs to loop through all available containers at each iteration and always try to get back onto the ideal path.  Penalties for deviation are totally flexible in that small deviations can be without penalty while sizable ones are some function that penalizes them heavily.  Other problem features can be captured, e.g., buffer space at a work center, and incorporated in the penalty function.  This is not a formal optimization, but it is speedy and good enough for the real world application.

 The watchwords seem to be these.  Look for the important features of a good solution.  See if a simple rule or concept will drive the solution toward these good features.  This is especially true when there is little or no economic benefit to the optimal solution.  Try to develop a heuristic that is flexible in adapting to normal variance in instances of the data and can be tuned to choose between competing objectives.

The summer issue of Manufacturing Today includes an article authored by Ted Schaefer and Alan Kosansky entitled Face Complexity – Making Sound Business Decisions.

“With every passing year, the amount and variety of information available to make business decisions continues its exponential growth. As a result, business leaders have an opportunity to exploit the possibilities inherent in this rich, but complex, stream of information. Alternatively, they can continue with the status quo, using only their good business sense and intuition and thereby risk being left in the dust by competitors. Top-tier companies have learned to harness the available data with powerful decision support tools to make fast, robust trade-offs across many competing priorities and business constraints.”

Read the complete article here: Face Complexity – Making Sound Business Decisions