In its broadest sense, health data refers to any type of data that provides use for improved research and innovation, as well as healthcare related decision making (Marjanovic et al. One example of such a data-rich ecosystem is global health, where the potential of data holds promise across all the building blocks of health systems. Indeed, problems plaguing data-rich ecosystems require understanding how the whole system will react to a seemingly inconsequential modification in one part of the system (Sterman 2001). These systems give rise to dynamic complexity because the system is: (1) constantly changing, (2) tightly coupled, (3) governed by feedback, (4) nonlinear, (5) history-dependent, (6) self-organizing, (7) adaptive, (8) characterized by trade-offs, (9) counterintuitive, and (10) policy resistant (Sterman 2001: 12). 1.1 Data-Rich Ecosystemsĭata-rich ecosystems are defined as “technological and social arrangements underpinning the environments in which is generated, analysed, shared and used” (Marjanovic et al. We will introduce, first, what we mean by data-rich ecosystems second, the terminology of system dynamics and third, a few applications of system dynamics in data-rich ecosystems. This section will proceed in three parts. The related systems modelling methodology of system dynamics involves computer simulation models that are fundamentally unique to each problem setting (Homer and Hirsch 2006: 452). Within system dynamics, causal loop diagrams are the main analytical tools that assist in the identification and visualization of key variables and the connections between them. If you like, you may add control group varying product life time, say from 0.5 to 10.System dynamics is a fundamentally interdisciplinary field of study that helps us understand complex systems and the sources of policy resistance in that system to be able to guide effective change (Sterman 2001). Discards mean there is always some fraction of the population in the potential adopter pool. The adoption rate rises, peaks, and falls to a rate that depends on the average life of the product and the parameters determining the adoption rate. Now, instead of falling to zero, the potential adopter population is constantly replenished as adopters discard the product and reenter the market. Observe the population dynamics using the chart. You can see that rate curves look exactly how we expected - the discard rate is actually the adoption rate delayed by 2 years - the life time of the product. Run the model and view plots for AdoptionRate and DiscardRate. You may check how the delay function works. Now we have finished modeling the product replacement purchases. The discard rate is null until the time of use of the first purchased products elapses. In our case, function reproduces AdoptionRate delayed on the ProductLifeTime value. The delay() function implements the time delay and has the following notation: Set the following formula for the flow variable:.You can see that stock formulas have changed as well:.Change the name of the flow to DiscardRate (do not forget to press Ctrl+Enter when finished typing new flow name). New flow from Adopters to PotentialAdopters is added.First, double-click Adopters stock in the graphical editor.So, the discard flow is nothing else but the adoption flow delayed on the average life time of the product.Ĭreate the discard flow Adopters to PotentialAdopters People move back from the adopter population to the pool of potential adopters when the product they have purchased is discarded or consumed. Assume that the average duration of active use of our product is 2 years.We will model repeat purchase behavior by assuming that adopters move back into the population of potential adopters when their first unit is discarded or consumed.įirst, we will define a constant representing the average life time of product. The model we have created does not capture situations where the product is consumed, discarded, or upgraded, all of which lead to repeat purchases.
0 Comments
Leave a Reply. |