US2014278507A1PendingUtilityA1
Methods and systems for growing and retaining the value of brand drugs by computer predictive model
Individually held — no corporate assignee on recordPriority: Mar 15, 2013Filed: Mar 14, 2014Published: Sep 18, 2014
Est. expiryMar 15, 2033(~6.6 yrs left)· nominal 20-yr term from priority
Inventors:Myrtle S. Potter
G06Q 30/0276G16H 70/40G06Q 30/0201G16H 20/10G16H 50/20G06Q 50/22
64
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
The present invention is directed to a brand value growth and retention system for brand drugs commercialized by brand drug advertisers through a brand drug's lifecycle during patent exclusivity and after loss of exclusivity. The brand value growth and retention system iteratively analyzes combined computational models of consumer, healthcare provider retailer and payor segment data to produce brand drug promotional campaigns that are predictive with modifying parameters that transform the promotional campaigns over time. As a result, the brand drug promotional campaign generates an increased number of brand drug purchases while predicting the point where incremental promotional campaign investments produce a diminishing number of incremental brand drug purchases.
Claims
exact text as granted — not AI-modifiedWhat is claimed and desired to be secured by Letters Patent of the United States is:
1 . A computer-implemented method for generating a promotional campaign in healthcare industry, comprising:
executing a first computational model on the consumer segment data to determine a first substantially optimal brand drug promotional mix for consumers who are candidates for a brand drug; executing a second computational model on healthcare provider segment data to determine a second substantially optimal brand drug promotional mix for healthcare providers who treat the consumers that are candidates for the brand drug; executing a third computational model on a computer model on retail store segment data to determine a substantially optimal product mix for retail stores that sell the brand drug; executing a fourth computational model on a computer model on payor segment data to determine a substantially optimal contracting strategy for the brand drug; and generating a promotional campaign for the brand drug by running a predictive model of the consumer segment data, healthcare provider segment data, retail store segment data and the payor segment data, based on the combination of outputs from the first, second, third and fourth computational models.
2 . The method of claim 1 , further comprising:
generating a first predictive element from the first computational model on the consumer segment data; generating a second predictive element from the second computational model on healthcare provider segment data; and generating a third predictive element from the third computational model on retail store segment data; generating a fourth predictive element from the fourth computational model on payor segment data; wherein the predictive model is generated based on a quadripartite combination of the first predictive element, the second predictive element, the third predictive element and the fourth predictive element.
3 . The method of claim 2 , wherein each of the first, second, third and fourth predictive elements partially affects the predictive model in generating the promotional campaign.
4 . The method of claim 1 , wherein the promotional campaign comprises a plurality of segment promotional plans, each promotional plan including one or more tactic profiles, each tactic profile being selected when a consumer segment in the consumer segment data responds to a particular promotional tactic.
5 . The method of claim 1 , wherein the promotional campaign comprises a plurality of segment promotional plans, each promotional plan including one or more tactic profiles, each tactic profile being created and selected when a consumer segment in the consumer segment data responds to a particular promotional tactic.
6 . The method of claim 1 , wherein the predictive model is adaptive to a change in a market response, the market response being affected by the first, second and third computational models.
7 . The method of claim 6 , wherein the predictive model is adapted via the application of a learning machine that estimates parameters thereby generating a transformed predictive model.
8 . The method of claim 6 , wherein the predictive model is adapted via the application of a learning machine that modifies existing parameters thereby generating a transformed predictive model.
9 . The method in claim 1 , wherein the predictive model combines information from computational models in a linear manner, wherein the combined information includes at least two of the consumer segment data, healthcare provider segment data, retail sales data, and the payor segment data.
10 . The method in claim 9 , wherein the combined information in the predictive model provides explicit weights to one or more components in the combined information.
11 . The method of claim 1 , wherein the predictive model is computed from the following equation:
SPP
=
∑
i
=
1
,
m
α
(
t
i
)
β
i
T
i
(
F
i
,
S
j
)
wherein the promotional campaign comprises a plurality of segment promotional plans (SPP), the symbol α(t i ) representing temporal preferences including order and weight, and the term β i *T i (F i , S j ) denoting one or more tactic profiles, coefficient β i denoting weighted factors applying respectively each corresponding tactic profile, the frequency F i applying to the corresponding promotional tactic T i up to the m th tactic, such that the symbol m represents the total number of tactics.
12 . The method of claim 11 , wherein the term α(t)'s denotes any ordering or parallelizing temporal function.
13 . The method in claim 1 wherein the promotional campaign is a weighted combination of the segment promotional plans (SPP's).
14 . The method of claim 13 , wherein the promotional campaign (PC) is represented by the following equation:
PC
=
∑
i
=
1
,
N
α
(
t
i
)
SPP
i
15 . The method of claim 1 , wherein the promotional campaign comprises explicit interaction terms among the actual or planned segment promotional plans as well as individual segment promotional plans (SPPs)
16 . The method of claim 15 , wherein the explicit interaction terms are binary, comprising a planned or actual SPP interacting with a second planned or actual SPP.
17 . The method of claim 1 , wherein the promotional campaign is represented by the following equation, where j=1, M ranges over all segment promotional plans in one or a plurality of promotional campaigns, including active or planned promotion campaigns, and the function G SPP (SPP i ,SPP j ) computes potential or actual interactions among a plurality of segment promotional plans contained in the promotional in the promotional campaign(s):
PC
=
∑
j
=
1
,
M
∑
i
=
1
,
N
α
(
t
i
)
[
SPP
i
G
spp
(
SPP
i
,
SPP
j
)
]
18 . The method of claim 2 , wherein the initial promotional campaign is modified to take into accounts interactions among the three predictive elements.
19 . The method of claim 18 wherein the interactions among the predictive elements are binary comprising a first predictive element interaction with a second predictive element.
20 . The method of claim 19 , wherein the interaction among the three predictive elements (PEs) is governed by the following equation, where the function G PE computes potential or actual interactions among the three predictive elements:
PC
modified
=
PC
initial
∑
j
=
1
,
4
∑
i
=
1
,
4
,
i
≠
j
[
1
+
G
pe
(
PE
i
,
PE
j
)
]
21 . The method of claim 1 , further comprising selecting a different segment promotional plan modifying the promotional campaign for a particular consumer segment within the consumer segments if the predictive model does not meet a predetermined substantially optimal threshold.
22 . The method of claim 21 , further comprising determining the benefit of a different promotional campaign and providing feedback to a learning machine to re-estimate parameters and revise corresponding predictions from one or more of the three predictive elements.
23 . The method of claim 1 , further comprising training a learning machine by invoking a machine learning method to estimate and attempt to optimize parameters for prediction of one or more computational models sourced from the executing step of the first computational model, the executing step of the second computational model, and executing step of the third computational model.
24 . The method of claim 23 , wherein the learning step comprises learning from data sourced from the current campaign, data sourced from at least one prior promotional campaign, and data sourced from external market reception to the current promotional campaign, to generalize learning from the collection of data.
25 . The method of claim 23 , wherein the machine learning method comprises an active or proactive learning in which a new tactic may attempt to jointly optimize both new knowledge gained about the effectiveness of the new tactics and the immediate impact of the selected campaigns and tactics.
26 . The method of claim 4 , wherein the machine learning method comprises an active or proactive learning in which a new promotional campaign attempts to jointly optimize both new knowledge gained about the effectiveness of the new promotional campaigns and the immediate impact of the selected promotional campaigns and promotional tactics.
27 . The method of claim 23 , further comprising re-estimating one or more computational models if not all results in the step of estimating and attempting to optimize parameters for prediction of one or more plurality of computational models are positive.
28 . The method in claim 23 , wherein the machine learning method suggests multiple potential but mutually exclusive improvements where one improvement is not positive and re-estimating the others from the feedback of the first tested improvement.
29 . A system for growing and retaining value in brand drugs, comprising:
a consumer segments module configured to execute a first computational model on the consumer segment data to determine a first substantially optimal brand drug promotional plan for consumers who are candidates for a brand drug; a healthcare provider module configured to execute a second computational model on healthcare provider segment data to determine a second substantially optimal prescription promotional plan for healthcare providers who treat the consumers that are candidates for the brand drug; a manufacturer payor module configured to execute a third computational model on payor segment data to determine a substantially optimal contracting strategy for the brand drug; and a financial model simulator module, coupled to the consumer segments module, the healthcare provider module, and the manufacturer payor module, configured to generate a promotional campaign for the brand-name drug by running a predictive model of the consumer segment data, healthcare provider segment data, and the payor segment data, based on the combination of outputs from the first, second and third computational models.
30 . A computer-implemented method for generating a promotional campaign in healthcare industry, comprising:
executing at least two computational models for two segment data to determine a first substantially optimal prescription promotional mix for a first segment data the and a second substantially optimal prescription promotional mix for a second segment data; and generating a promotional campaign for the brand drug by running a predictive model of the first segment data and the second segment data, based on the combination of outputs from the first and second computational models.Join the waitlist — get patent alerts
Track US2014278507A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.