Table 3.

Number of articles reviewed addressing each category

			2017–2019	2007–2016
Resilience characteristics	Plan	Uses NAS phase of resilience	65	41
	Absorb	Uses NAS phase of resilience	67	44
	Recover	Uses NAS phase of resilience	72	31
	Adapt	Uses NAS phase of resilience	54	17
	Resilience Metric or “Proxy”	Numeric measure of resilience, separate from network characteristica	45	31
Supply chain model representation	Linear	Unidirectional and no path options	10	10
	Branching	Unidirectional with path options	29	15
	Graph	Multi-directional with path options	38	17
	Otherb	Different enough to not classify as the others	17	5
Transportation model representation	None	Disruptions only within SC nodes	50	18
	Same as SC	SC links can be disrupted	10	12
	Independent Links	Independent network for each SC link	28	10
	Graph	Can be adjusted dynamically—SC nodes placed on a single transportation network	5	5
	Other	Different enough to not classify as the others	2	2
Command and control representation	None/Pre-determined	Decisions on production and movement will not change	20	4
	If–then/Heuristic	Discrete, deterministic and simple rules for decision at each SC node; explicit guidance for managerial implications presented	18	5
	Agent Based	Probabilistic and/or complex made by multiple independent actors	13	5
	Optimization	Optimization algorithm employed to ensure "best" decisions made	38	31
	Other	Different enough to not classify as the others	6	2
Disruption representation	Case Study	Real world circumstances used to model disturbances	44	15
	Set List	Pre-determined list of disturbances generated	10	26
	Monte Carlo	Disturbances randomly generated	11	7
	Targeted	Adversarial algorithm used to generate disturbances	5	1
	None/Other	Different enough to not classify as the others	25	4

^aIn the 2017–2019 review, “proxy” was added as a way to differentiate between the numeric measure of supply chain resilience and other factors being used to measure supply chain resilience (e.g., disruption cost, product depreciation, time to receive a good, etc.). Of the 45 publications shown to contain a metric, 24 are considered proxy

^bIn the 2017–2019 review, “other” encompasses supply chain models that focus more on resilience of the network within external contexts, as opposed to strict definitions and traditional graphic models of supply chains. Of the 17 publications considered other, 9 are considered to model supply chain resilience in relation to other networks, rather than within itself