2-3 December 2009, London. Institution of Mechanical Engineers, Engineering in Medicine and Healt... more 2-3 December 2009, London. Institution of Mechanical Engineers, Engineering in Medicine and Health Division.
As documented in the recent OECD report &... more As documented in the recent OECD report 'the adverse outcome pathway for skin sensitisation initiated by covalent binding to proteins' (OECD, 2012), the chemical and biological events driving the induction of human skin sensitisation have been investigated for many years and are now well understood. Several non-animal test methods have been developed to predict sensitiser potential by measuring the impact of chemical sensitisers on these key events (Adler et al., 2011; Maxwell et al., 2011); however our ability to use these non-animal datasets for risk assessment decision-making (i.e. to establish a safe level of human exposure for a sensitising chemical) remains limited and a more mechanistic approach to data integration is required to address this challenge. Informed by our previous efforts to model the induction of skin sensitisation (Maxwell and MacKay, 2008) we are now developing two mathematical models ('total haptenated protein' model and 'CD8(+) T cell response' model) that will be linked to provide predictions of the human CD8(+) T cell response for a defined skin exposure to a sensitising chemical. Mathematical model development is underpinned by focussed clinical or human-relevant research activities designed to inform/challenge model predictions whilst also increasing our fundamental understanding of human skin sensitisation. With this approach, we aim to quantify the relationship between…
This article illustrates the use of multi-agent modelling and prediction of consumer goods market... more This article illustrates the use of multi-agent modelling and prediction of consumer goods markets. A behavioral model incorporating utility based rational choice enhanced with psychological drivers is presented to study a typical market, characterized by repeat purchase incidences by households. The psychological drivers incorporate purchase strategies of loyalty and change-of-pace, which affect the choice set of consumer agents in an agent based simulation environment. Agent specific memories of past purchases drive these strategies, while attribute specific preferences and prices drive the utility based choice function. Transactions data from a category in a supermarket is used to initialize, calibrate and test the accuracy of predictions of the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increasing the memory length beyond a certain limit does not improve pr...
The volatility in a CPG market is modeled using a bottom-up simulation approach and validated aga... more The volatility in a CPG market is modeled using a bottom-up simulation approach and validated against disaggregated supermarket transactions data. The simulation uses independent agents, each agent representing unique households in the data. A simple behavioral model incorporates household preferences for product attributes and prices. Our validation strategy tests the model predictions at both macro and micro levels and benchmarks the performance in each against a random choice model. The model significantly outperforms the benchmark at both levels. At the macro level, choices made by heterogeneous agents accurately captures the volatility in market shares over time. This accuracy at the macro level is driven by the accuracy of predictions at the micro household level SKU and attribute choice.
An agent based behavioral model incorporating utility based rational choice enhanced with psychol... more An agent based behavioral model incorporating utility based rational choice enhanced with psychological drivers is presented to study a typical consumer market. The psychological drivers incorporate purchase strategies of loyalty and change-of-pace, using agent specific memory of past purchases. Attribute specific preferences and prices drive the utility based choice function. Transactions data is used to calibrate and test the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increased agent memory does not improve predictions in the model beyond a threshold, indicating that consumer memory of past shopping instances is finite and recent purchase history is more relevant to current decision making than the distant past. The article illustrates the use of agent based simulations to model changes or interventions in the market, such as new product introductions, for which no past history exists.
2-3 December 2009, London. Institution of Mechanical Engineers, Engineering in Medicine and Healt... more 2-3 December 2009, London. Institution of Mechanical Engineers, Engineering in Medicine and Health Division.
As documented in the recent OECD report &... more As documented in the recent OECD report 'the adverse outcome pathway for skin sensitisation initiated by covalent binding to proteins' (OECD, 2012), the chemical and biological events driving the induction of human skin sensitisation have been investigated for many years and are now well understood. Several non-animal test methods have been developed to predict sensitiser potential by measuring the impact of chemical sensitisers on these key events (Adler et al., 2011; Maxwell et al., 2011); however our ability to use these non-animal datasets for risk assessment decision-making (i.e. to establish a safe level of human exposure for a sensitising chemical) remains limited and a more mechanistic approach to data integration is required to address this challenge. Informed by our previous efforts to model the induction of skin sensitisation (Maxwell and MacKay, 2008) we are now developing two mathematical models ('total haptenated protein' model and 'CD8(+) T cell response' model) that will be linked to provide predictions of the human CD8(+) T cell response for a defined skin exposure to a sensitising chemical. Mathematical model development is underpinned by focussed clinical or human-relevant research activities designed to inform/challenge model predictions whilst also increasing our fundamental understanding of human skin sensitisation. With this approach, we aim to quantify the relationship between…
This article illustrates the use of multi-agent modelling and prediction of consumer goods market... more This article illustrates the use of multi-agent modelling and prediction of consumer goods markets. A behavioral model incorporating utility based rational choice enhanced with psychological drivers is presented to study a typical market, characterized by repeat purchase incidences by households. The psychological drivers incorporate purchase strategies of loyalty and change-of-pace, which affect the choice set of consumer agents in an agent based simulation environment. Agent specific memories of past purchases drive these strategies, while attribute specific preferences and prices drive the utility based choice function. Transactions data from a category in a supermarket is used to initialize, calibrate and test the accuracy of predictions of the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increasing the memory length beyond a certain limit does not improve pr...
The volatility in a CPG market is modeled using a bottom-up simulation approach and validated aga... more The volatility in a CPG market is modeled using a bottom-up simulation approach and validated against disaggregated supermarket transactions data. The simulation uses independent agents, each agent representing unique households in the data. A simple behavioral model incorporates household preferences for product attributes and prices. Our validation strategy tests the model predictions at both macro and micro levels and benchmarks the performance in each against a random choice model. The model significantly outperforms the benchmark at both levels. At the macro level, choices made by heterogeneous agents accurately captures the volatility in market shares over time. This accuracy at the macro level is driven by the accuracy of predictions at the micro household level SKU and attribute choice.
An agent based behavioral model incorporating utility based rational choice enhanced with psychol... more An agent based behavioral model incorporating utility based rational choice enhanced with psychological drivers is presented to study a typical consumer market. The psychological drivers incorporate purchase strategies of loyalty and change-of-pace, using agent specific memory of past purchases. Attribute specific preferences and prices drive the utility based choice function. Transactions data is used to calibrate and test the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increased agent memory does not improve predictions in the model beyond a threshold, indicating that consumer memory of past shopping instances is finite and recent purchase history is more relevant to current decision making than the distant past. The article illustrates the use of agent based simulations to model changes or interventions in the market, such as new product introductions, for which no past history exists.
Uploads
Papers by Stephen E Glavin
micro household level SKU and attribute choice.
calibrate and test the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increased agent memory does not improve predictions in the model beyond a threshold, indicating that consumer memory of past shopping instances is finite and recent purchase history is more relevant to current decision
making than the distant past. The article illustrates the use of agent based simulations to model changes or interventions in the market, such as new product introductions, for which no past history exists.
micro household level SKU and attribute choice.
calibrate and test the model. Results indicate that prediction accuracy at both macro and micro levels can be significantly improved with the incorporation of purchase strategies. Moreover, increased agent memory does not improve predictions in the model beyond a threshold, indicating that consumer memory of past shopping instances is finite and recent purchase history is more relevant to current decision
making than the distant past. The article illustrates the use of agent based simulations to model changes or interventions in the market, such as new product introductions, for which no past history exists.