In my article in the July/August 2021 issue of PARCEL (“Is It Time to Move to a New Warehouse?”), I explained how you can determine whether it might make sense to move your fulfillment operation to a new warehouse. Once you’ve decided it might be time to move, how do you decide where it should be located so it will provide the best service levels for the lowest operating costs?
This follow-up article outlines an objective, emotion-free process for determining where to locate your new warehouse. We’ll set aside some of the more obvious activities, like getting the right team together, interviewing key company stakeholders, etc., and focus on the process itself. Also, we’ll look at optimizing the location of a single warehouse. (Optimizing the locations of multiple warehouses in a logistics network follows a similar process, but it can add significant complexity due to the proliferation of the variables involved.)
Here are seven recommended steps for objectively determining the best location for your new warehouse:
1: Define Your Business Requirements Over Your Planning Horizon
The first step of the age-old engineering method is to define the problem. That philosophy also applies to determining where your warehouse should be located. Start by identifying your planning horizon for your study — three to five years from now is typical — and forecast the requirements of your warehouse over that time. For example, define the following:
·Sales volumes – What will your projected shipping volumes be, in units (not dollars), during the average week, as well as during the week when you experience your seasonal peak volume?
·SKUs – What SKUs will you be receiving, storing, picking, and shipping? What are their attributes?
·Customer orders – What types of orders will you be fulfilling? What are their attributes, and how will you ship them?
·On-hand inventory – What will your projected maximum on-hand inventories be, in units (not dollars), during the average week, as well as during the week when you experience your seasonal inventory peak?
·Suppliers – Who will your suppliers be? How will your warehouse need to accommodate inbound shipments from them?
·Customers – More importantly, who will your customers be? How will your warehouse need to accommodate outbound orders and meet their expectations, such as their desired service levels?
2: Develop an Objective Scorecard
Since you will be evaluating several alternative warehouse locations, your next step is to decide how you will measure and evaluate your alternatives objectively. Make a list of all the data points and attributes that you’ll want to compare among the alternatives. Here are some common examples:
·Annual transportation costs (inbound and outbound freight costs)
·Annual facility costs (rent, purchase costs [principal, interest, etc.], depreciation, utilities costs, insurance costs, taxes)
·Annual equipment and technology costs (manual and mechanized equipment costs, automation costs, costs for computers and related devices, software costs, maintenance costs)
·Loaded annual full-time, part-time, and temporary labor costs, including benefits and overtime
·Initial warehouse relocation costs
oAverage supplier (inbound) delivery time per order, in days
oAverage customer (outbound) delivery time per order, in days
·Any intangible and qualitative factors that are important to your business
Keep in mind that the more data points and attributes you document on your scorecard for comparison, the more comprehensive your comparison will be. On the other hand, more points of comparison will require more time and effort during your evaluation. Don’t create unnecessary work, and be sensible about determining which factors will be the most important for your business.
3: Select the Right (Optimization) Tool for the Job
Accurate optimization of a warehouse location also requires an effective modeling and optimization tool to analyze your supply chain data, so your next step is to select a good one for your business requirements. The right tool for this purpose will help you identify the best location for your warehouse over your planning horizon.
Notice that I said the “right” tool, not the “best” tool. A variety of modeling and optimization tools are available to you. They range from ubiquitous spreadsheet applications that you probably already use to highly sophisticated software programs that can be leased by subscription for upwards of $10,000 per month.
How good is good enough? The right modeling and optimization tool for you depends on your desired level of comprehensiveness and accuracy and the size and complexity of your warehousing operations. There’s no need to cut butter with a chainsaw. Often the “best” optimization tool is overkill for the needs of a business, and it will almost always add more cost, time, and complexity to the process.
If you plan to model your warehousing operations in-house, discuss your business requirements with several reputable developers of modeling and optimization tools to determine which solution suits your needs. On the other hand, if you hire a good logistics consultant to analyze your warehousing operations, they can probably provide the right modeling and optimization tool for your business requirements as part of their project engagement.
4: Gather Data, Data, and More Data
Make no mistake about it: Determining the optimal location of a warehouse is an intensely data-driven exercise, and the GIGO rule (garbage in, garbage out) profoundly applies to this process. I remind my clients of this fact so much that I sound like a broken record, and rightfully so. It’s a universal truth that is too often ignored.
Consequently, gathering comprehensive, accurate supply chain data for each item on your scorecard is your next step. Data gathering, “scrubbing,” consolidation, and input (into your planning and optimization tool) will, without a doubt, be the most time-consuming and tedious activities in the process. But if you don’t model your alternatives using good data, there’s no point in doing it at all.
5: Accurately Model Your Existing Warehouse Location
Then the fun begins! You’ll need to model the projected costs and service levels of your current warehouse, so you have something to compare with those of any other potential warehouse locations you decide to consider. In this step, configure your current warehouse operations into the modeling and optimization tool you selected and make sure your model is accurate. This initial model will serve as your “base-case” (baseline) model, and it should realistically represent your current warehousing operations, both historically and over your planning horizon (assuming you stay put and don’t move to a new warehouse).
To configure this base-case model in your tool, set up the locations and relevant attributes of your current warehousing operations, including:
·Your current warehouse
·All your suppliers’ ship-from locations
·All your customers’ delivery (demand) locations
·Any other physical locations in your distribution network (ports, pool points, cross-dock facilities, etc.)
Then import at least one year’s worth of your historical data (outlined in the sections above) and freight-rate information into your tool to see if the calculated costs and delivery times match your actual historical costs and delivery times. If the calculations of costs and service levels aren’t reasonably close to your actual historical costs and service levels, then your model isn’t accurate, so you’ll need to correct your model. If they are reasonably close, that’s great news — you’ll know your model is valid. Now you can update your data set with the numbers you projected over your planning horizon, and you’ll be ready to model your alternatives.
6: Define and Model Your Alternative Warehouse Locations
This step involves deciding which alternative warehouse locations you want to configure into your modeling and optimization tool and evaluating them. Apply your team’s collective logistics expertise to decide which potential warehouse locations are worth considering and comparing. Depending on which modeling and optimization tool you’re using, your software might also be capable of suggesting optimal warehouse locations.
Assign a unique name for each alternative warehouse location you decide to model. For example, I typically call the base case “Alternative 0” and number all the other potential alternatives accordingly (Alternative 1, Alternative 2, etc.). Since the data set in your modeling and optimization tool was updated in the previous step to include your projections over your planning horizon, just replace the warehouse location attributes in your tool for each alternative, and save each alternative’s model under the name you assigned to it.
Whether you’re looking for an existing building or a greenfield site, before you lock down the potential warehouse location for each of your alternatives, it’s a good idea to do some sensitivity testing in your models. Sensitivity testing is a fancy name for doing some trial and error in your models, and this is usually very easy with modeling and optimization software. You can slightly tweak the potential location of the warehouse in each model to see what effect that has on the projected costs, service levels, and other factors. For example, you might move your potential warehouse into a known industrial area, or into a county or city with lower taxes, or to a location with easy access to major highways, or to a location more convenient to places that might appeal to a future warehouse workforce (residential areas, public transportation, daycare, restaurants, etc.).
7: Objectively Compare All Alternatives and Select the Winner
Once you have finalized all your alternative warehouse locations and finished running each of their models using the same data set for your planning horizon, your final step is to compare your alternatives and select the best warehouse location for your business. You can easily achieve this by creating a simple spreadsheet table. List all your alternative warehouse locations (including your base-case alternative) across the columns at the top, and list all the data points and attributes you defined for your scorecard down the left-hand column. Then enter the resulting metrics into your modeling and optimization tool calculated for each alternative into its respective column.
Voila! This table will give you a proforma, apples-to-apples, summarized snapshot of the costs, service levels, and other intangible and qualitative factors associated with each alternative warehouse location, without letting your emotions influence your thinking. You will objectively and easily see which warehouse location will deliver the greatest value for your business, and by how much. Then it will be time to start planning that new warehouse to turn those projected benefits into a reality!
Stephen T. Hopper, PE is Founder & Principal of Inviscid Consulting, whose mission is to help business plan and streamline their warehousing, logistics, manufacturing, and distribution operations to drive down operating costs, boost capacity, improve service levels, and mitigate risk. He can be reached at firstname.lastname@example.org or 404.832.5326.
This article originally appeared in the September/October, 2021 issue of PARCEL.