The Parcel Revolution (2 of 4): Network Resilience

This series of posts is designed to discuss some of the challenges faced by mid-sized postal organizations and parcel carriers wanting to take advantage of the growth in parcel volumes.  As we mentioned in the initial post, it is critical to develop a clear operations strategy focused on parcel distribution, rather than “piggy-backing” parcels on the letter mail processing network.   How many hubs? How far apart? What processing windows?  These are just a few of the key questions that must be answered when designing the right parcel distribution network.  Overnight products leave very little time for processing, requiring high-capacity, high-speed equipment; two- or three-day products allow a more efficient design of the distribution network, of hubs and of terminals.

comScore recently forecasted that 2014 e-Commerce sales in the U.S. will grow 16% over 2013.  At the same time, major shippers like Amazon are creating new distribution centers, changing distribution patterns and increasing the demand for shorter, regional trips.  There is no point in designing a parcel network that achieves optimal efficiency if the first change causes the system to collapse. The carrier’s distribution network must be designed to adapt to changing distribution patterns and to other changes in environmental conditions (also referred to as environmental requests).

Resilience is defined as the ability to maintain critical functionality in the face of significant stress. Resilient networks (also said to be”robust”) exhibit two key abilities: infrastructure capacity to respond (equipment, people) and a response culture.  More about the latter in another post.  Defining the optimal level of infrastructure needed to maintain robustness is not trivial.  Interestingly, the study of complex adaptive biological systems provides some answers (1) (2).  These complex systems usually show two characteristics:

• Partial redundancy, or the ability of the network’s nodes and links to (partially) take over a similar function in other parts of the network.  This means, for instance,  the ability to process, sort and transport on behalf of other parts. This way, if a part of the network is lost or overwhelmed, other parts can help.
• Versatility is the ability of network nodes and links to (partially) perform other functions than their primary role.  For instance, primary sort hubs should be able to provide collection or delivery for other terminals in case of need.  This capability has also been referred to as “functional plasticity” in the study of complex adaptable systems.

Planning such a network is now highly doable.  The robustness of a planned network of hubs and links can be tested in response to environmental changes by using a simulation model such as the one developed to study USPS consolidation here and here.  Various levels of redundancy and versatility can be tested and optimal levels can be selected.  Of course, as mentioned above, infrastructure is only half of the equation, a response culture is also necessary.

(1) Whitacre J. M., “Degeneracy: a link between evolvability, robustness and complexity in biological systems”  Theoretical Biology and Medical Modelling, vol. 7(6), 2010
(2) Braendle, C. and M.A. Félix, Plasticity and Errors of a Robust Developmental System in Different Environments. Developmental Cell, 2008. 15(5): p. 714-724.