In Part I, we defined Supply (Chain) Network Optimization as the optimization of your global distribution network to minimize costs while controlling risk to an acceptable level, discussed the various costs involved in a supply chain network, outlined some of the questions you should be asking, and outlined some of the complexities associated with Supply Chain Network Optimization.

In Part II, we reminded you that supply network optimization, of which freight / transportation cost optimization is a significant part, does not belong as part of a standard sourcing project (since only freight rates, and not total costs, can be fixed and freight costs will always be an approximation anyway) and that if you are considering an award where freight dominates the cost, then you should be optimizing your transportation cost and estimating your unit cost (based upon quoted rates and expected discounts from pre-qualified suppliers in the region), since the largest savings are generally achieved from optimizing the component of spend that makes up the majority of spend, not the minority.

Today, we going to discuss some of the challenges of global distribution, as highlighted in the ESYNC white-paper, “Strategic Assessments” (registration required) that prompted me to white this series of posts. Even though the white paper itself did not live up to the promises implied by its title (it’s sorely lacking on the distribution network front and the strategy front), the Operation’s Analysis & Opportunity Assessment table on page 2 and 3 did a good job summarizing the challenges with (global) distribution and outlining why you need appropriate supply chain information technology to help you manage your global supply network.

The table also included an “ESYNC Approach” column which, predictably, revolved around AIDC, WMS (Warehouse Management Systems), TMS (Transportation Management Systems), and RFID – technologies that ESYNC sells or integrates with. These are good suggestions, but they don’t completely address the problem. You need IMS (Inventory Management Systems) that go beyond the warehouse and integrate with forecasting systems; GDM/GTM (Global Data Management / Global Trade Management) systems that help you comply with all of the regulatory requirements of the countries you operate in, all of the import and export customs requirements, and all of the taxation laws (including those that allow you to claim refunds under certain conditions); and proper modeling and optimization tools to make sure you have the right network in the first place.

Yesterday, I said that Supply (Chain) Network Optimization (SCNO, or SNO, but not snow), which is not easy, is the optimization of your global distribution network to minimize costs while controlling risk to an acceptable level. Furthermore, it’s not something you should tackle as part of your everyday sourcing project.

Yes, I’m saying it again. And the next person who dares to suggest otherwise gets an e-coupon for a free smack upside the head, redeemable the next time we meet. The nature of distribution network optimization is that it cannot be optimized within a single sourcing scenario, and any attempt to do so is likely to do more harm than good. To truly optimize your network, you have to optimize across all of your buys, and even in any given year, you’re likely renegotiating less than a third of your major contracts and a quarter of your buys, and you don’t expand into new countries overnight.

Furthermore, since you are contracting for fixed rates, not fixed costs, and your total costs depend on actual volumes, which are forecasted and variable, your freight costs at the time of award are approximations. If you’ve done your homework, and are revisiting your freight contracts semi-annually or annually, like you should be, you should be within a few percentage points, at most, with these estimates. Let’s say worst case scenario where you’re off by 5% because you didn’t quite meet demand to get that rebate you were hoping for (since your estimate demand was only 1% above the cutoff). In an average strategic sourcing scenario, you’re looking at freight in the 5% to 15% range, so you’re off by at most 5% of 15% or 0.75%. Worst case, you’re off by 0.75% – but since you have an equal chance of being off either way if you’re estimating properly, it’s going to average out over all your awards and you’ll be within a fraction of a percent of optimal (unless oil spikes again and huge surcharges come into play, but since all your competitors will be paying the same surcharge that every carrier will be levying above the board, you’ll still be ahead compared to your peers, relatively speaking, if you have the right lanes at the right volumes for the right rates with the right carriers).

The proper way to optimize distribution network costs is to semi-annually or annually analyze all of your projected transportation needs over the next 6 to 12 months using all of your projected shipments (based upon current contracts, forecasts, and current patterns), aggregate volumes across lane groups (defined as the set of lanes that take a product from region A (such as a set of posts on the southwest coast) to region B (your major re-distribution center outside Chicago), bid out the appropriate lanes or lane groups to one or more carriers, and optimize a transportation award to these carriers who quote rates based upon minimum volume guarantees (such as 75% of expected volume across the lane). This gives you highly accurate freight rates to use in your award allocations.

On top of this, every one to three years you should be re-optimizing the flexible aspects of your distribution network. By this I mean, re-evaluating warehousing space that you are leasing or that is highly liquid and could be easily sold, re-evaluating the air and ocean freight options to you, re-evaluating the ports you are using, and re-evaluating your shipment consolidation strategy (should you always wait for shipments from multiple suppliers to fill the container or should you use a third party that can consolidate shipments for multiple buyers to fill the container).

For example, maybe sales on the west-coast are high and the east-coast are low, maybe you’ve experienced three major delays at your current port of choice, and maybe part of your cargo is always depreciating by waiting for enough cargo to fill a container. In this case, you would model the expected costs of retaining your current distribution network versus the costs of different network designs that cancelled the lease of one of your east coast distribution facilities, used a different port, and / or used a third party logistics provider to determine whether or not a redesigned network could offer you a lower operating cost (taking into account the one-time fixed costs associated with any restructuring you had to do, amortized over appropriate minimum contract periods).

Then, every three to five years, you should be re-optimizing the distribution network as a whole, including the fixed aspects of your distribution network. By this I mean every single warehouse and facility, every single warehouse and distribution facility, every single port of exit and entry, and every single carrier is considered to be floating and the cost of the current network is compared against a fully optimized solution (that takes into account all of the one-time fixed costs that will be associated with any and all restructuring) optimized over different, appropriate, minimum contract periods (from three to seven years, as the optimal model may change from year to year as volume projections will change year over year and existing contracts will expire, decreasing the fixed costs associated with a restructuring). Then, each of the optimal network model solutions that represent a significant cost savings should be compared against the current network model and those that represent the minimal deviations from the current model analyzed in depth. (Even if a totally redesigned network could offer lower cost savings, it’s a bad idea to re-design your whole network at once. Most big-bang efforts end in failure. Furthermore, it usually only takes a few major changes to produce significant savings, if done, right, and the longer it takes to redesign your network, the more you are relying on the accuracy of your long term forecasts to realize these savings, increasing the risks associated with redesign.

Finally, at least the first few times you do this, be sure to bring in an expert and make sure the appropriate tools are available. Even though these resources may be expensive, partly due to the relative lack of expertise in this area and partly due to the significant education and experience required to do these projects properly, (since you need expertise in Operations Research (OR), optimization, modeling, supply chain, analysis, and consulting), as another resident guru of the blog-sphere would be quick to point out, it’s not what you pay, but what you get out of it. You might think that 3K to 5K a day is expensive, but if that person, with access to the right tools, saves you millions of dollars, a hundred grand or two in professional services, (on-demand) software costs, and project costs is paltry in comparison. It’s the net value of the ROI, not the cost, that’s important, as it is in any business project.

Reading a recent white-paper by ESYNC entitled “Strategic Assessments” (registration required), I realized that even though I blog regularly about the different aspects of supply chain optimization, including supply chain network optimization (SCNO, or SNO, but either way us northerners will think you’re talking about the cold wet white stuff [different from The White Stuff] if you say it and don’t spell it), I have’t really tackled it end-to-end all in one series. So, in this post, I’m going to try and outline what’s involved in end-to-end supply network optimization.

Whereas sourcing optimization, the topic I tend to blog about the most, tends to focus on product or service award optimization from a total value perspective, where the transportation costs are assumed relatively fixed based upon supplier or third party carrier bids (or historical costs with adjustments to account for inflation and expected financial outlooks), supply network optimization tends to focus on optimization of the distribution network, and the relatively fixed transportation costs it entails.

As I discussed previously in my CombineNet (acquired by Jaggaer) series [you know, the guys who can Optimize Anything], the costs associated with transporting your goods from a supplier’s plant to an end customer depend on more than just where you’re sourcing from and where you’re sourcing to, but depend on the routes you’re using, the number of times your product needs to be on-loaded and off-loaded, the number of different carriers you are using, and the amount of time you have to pay for your product to sit in a (temporary) warehousing facility.

The costs involved in that average price of approximately $1.00 / kg per lane that you use in your decision optimization to determine a supplier award for a category can be quite numerous, especially when sourcing from China. There’s the transportation costs associated with the carrier that takes your cargo to a warehouse on the docks, the storage costs while you wait for shipments to consolidate from your various suppliers, the loading costs onto the barge, the export tariffs and duties to get the product out, the port fees, the ocean freight shipping costs, the port fees on the other side, the import tariffs and duties, the costs associated with loading and shipping the product with your local carrier to your primary distribution center, the cost of maintaining the local distribution center, and the costs of finally shipping the product to the retailer or end consumer, to name a few.

But Supply Network Optimization is about more than just optimizing these costs, it’s optimizing your network to allow for these costs to be optimized to a new global minimum, which is dependent on your network, while controlling risk. Should you have five central distribution centers in the US or ten? Should you use air or rail? Should you ship to one coast or both? Should you use one port or two? Should you ship by truck or rail? Should you handle your own logistics or outsource it to a 3PL? What’s the optimal number of carriers? And so on.

Not an easy problem, and not one any single vendor or consultancy is necessarily capable of advising on across the board (but I’m sure we’ll be hearing from Paul Martyn shortly about how CombineNet’s optimizer, in the right hands, can handle this and so much more), but not one you can ignore – and not one you should tackle in a sourcing project.