Network Analysis

Why it needs attention: Every company’s supply chain is in constant flux.  Customers change product lines, grow, move, or demand different services.  Storage requirements change, leases expire, and value added services may become a requirement.  Vendors shift sources, change product configuration, order minimums and delivery conditions.  And freight companies appear and disappear, fuel costs change, and new ports and hubs open.  Companies merge and divest. Each of these is an imposed change that affects the cost of the supply chain, and how competitively and profitably your company functions.

What we can do: Our team has worked together for over 17 years on network studies for pharmaceutical companies, national and regional mass merchants, and industrial product manufacturers and distributions to assess to alternatives available and the cost and service impact of changes in their network.  We use a commercially available network model along with our own enhancements to create a flexible, responsive and accurate tool that our clients use and re-use whenever their business changes.  Our system allows scenarios for various changes in volume, sources and customers.  We can limit facility sizes, delivery times and product lines.  And we can make changes to any or all of these factors in each evaluation run.  Distribution center location modeling techniques answer the following basic questions:

  • How many warehouses should there be in the distribution network?
  • Where should these warehouses be?
  • What should be the storage and throughput capacity of these warehouses?
  • How should products be allocated to the warehouses?
  • How can a move be made from the existing network to an improved network?

The Process: Our process rigorously follows a logical progression to arrive at a supportable conclusion.

Step 1: Data Collection:  Transportation costs, quantities, geographic locations, and product groups must be gathered for both  (1) supply data (shipments to warehouses) and (2) demand data (warehouse shipments to customers).   Warehouse data (for existing warehouses and candidate sites, if available) must include both the  fixed costs (lease or depreciation, site fees, taxes, systems, security) and variable costs (Labor, HLP, supplies).  In addition, forecasts by geographic region and product group, service zone limits, inventory carrying costs, building lease expirations, expansion limitations, capacity restrictions, allowable stock-out percentages, and other factors need to be documented for analysis.

Step 2: Baseline Validation:  This step entails setting the model’s parameters for the initial network validation state, using the location modeling software, to approximate existing distribution network costs and capacities.  This model must be validated against existing conditions to confirm that the model is accurately representing the actual real-world network, within an acceptable margin of error.  Once the model is developed and validated, alternative scenarios can be programmed.

Step 3: Alternative Generation:  Once the baseline is set, we can proceed to modify both the demand and source data to reflex future business requirements.  Because our model is site neutral, we do not artificially limit the potential DC locations.  Rather our model takes as input the number of DC location to be considered in the network and selects the least cost/best service locations.  By altering the number of DCs in each scenario, we can identify the best number and location set.  If desired, the model can be restricted in such a way as to require a DC in one or more locations and establish maximum capacities for each of the required locations

Step 4: Optimization and Refinement:  For each alternative defined, the model projects the key comparative data: (1) inbound costs from each supply point to each warehouse, (2) outbound costs from each warehouse to each customer, (3) a determination of which warehouse services each customer, (4) cost of inventory per warehouse, and (5) the product group to be handled at the facility.  Overall costs, service levels, existing conditions, and available sites in the selected region will then be used to determine the ideal warehouse locations.

A typical network modeling study will take from a few weeks to several months, depending on the complexity of the network, the availability of data, and the experience and skills of the project team.

Conclusion: Many factors go into an assessment of a distribution network.  Unfortunately, many of these factors come into conflict with each other, which adds to the complexity of the decision making process.  For instance, additional warehouses closer to customers will decrease delivery costs, but will increase inbound and inventory holding costs.

The Jack Kuchta LLC Team has the expertise and experience to model the distribution network in order to properly recommend a distribution network.  We can model a wealth of alternatives at a low marginal cost in order to give our clients a complete understanding of the cost factors affecting their distribution network and to provide enough options to find affordable warehouse space that can accommodate an optimized network design.

If you would like more information on how Jack Kuchta LLC can help you with your distribution network design or any of our range of services, please do not hesitate to call 201-579-8822 or email us at