An Agent-Based Model for Quantifying the Economic Value of Geographic Information
The aim of this thesis is to establish the market economic framework for geographic data exchange. We are interested in economic aspects of geographic information and the value it has for the potential buyer. Our hypothesis is that it is possible to quantify the value of geographic information in a specific decision making situation. Our general approach for verifying the hypothesis and testing the agent-based computational model is applied to a car navigation case study. The major scientific contributions of this thesis are a formal agent-based model, a conceptual model of dataset qualities that link properties of a dataset to the cost of the decision making process and results of simulations which give a quantitative value of geographic information in a particular decision making process. In the final sections of this chapter we present the scientific results, the intended audience of this research and an overview on the content of the subsequent chapters.
‘A cynic is someone who knows the price of everything but the value of nothing.’
There is a difference between the price at which something can be bought and the real value the people attach to it. A painting by, say Klimt, Picasso or Dali may be priced at millions of Euros and, whereas some people would buy it because they see it as an investment, others would buy it because they actually like the picture. Still others would look at it as just another picture, which could have been painted by anyone, and therefore regard it as overpriced.
Imagine that someone needs geographic information that can be acquired from a geographic dataset. What are the properties of a dataset that have economic value for the potential buyer? How much would the potential buyer be willing to pay for such information?
Pricing is a critical element of marketing geographic datasets. A price set too high is a barrier to access the data and prevents the widespread use of the data. A price set too low will not cover the high costs of collecting and maintaining the dataset and will not provide the expected rate of return on the producer’s investment (Krek and Frank 1999).
The characteristics of geographic data make application of the standard pricing theory difficult. Geographic data is not a standard economic good. The same dataset can be exchanged and used many times; the consumption by one user does not diminish the amount available for other users. New products can be easily created by forming different combinations of datasets. Geographic data has characteristics of an ‘experienced’ good as opposed to the ‘search’ good. The main characteristic of a search good is that consumers can value the quality of a good before actually using and testing it. A good is said to be an experienced good if consumers must experience a good in order to value it. The quality of an experienced good can be established only after actually purchasing it (Nelson 1970). Virtually any product that is new on the market has characteristics of an experienced good. Producers have developed strategies such as free samples, promotional pricing, and test versions to help potential buyers to learn about new goods. Information is an experienced good every time it is consumed. A consumer cannot know whether today’s Wall Street Journal is worth 75 cents until he has read it (Shapiro and Varian 1999). The main issue for the buyer of a search good is to select the product, and for the buyer of an experienced good it is how to learn about the product quality and its value.
In order to be able to set a price for the geographic dataset the data producers have to know what are the properties of the dataset that potential buyers value and how much do they value them. The price paid is a measure of the value that buyers place on the products. Understanding of concepts of value and the value formation will help the data producers to set a profitable and competitive price for their products. This approach naturally leads to value pricing where the buyer pays the price for the product according to the value he attaches to this product.
The only economic use of the geographic dataset is to support and improve the decision making process of the user. Only the data directly related to the improvement of the decision has an economic value for the user (Frank 1996). In order to be able to assess the price for an exchange the producer and the buyer of a dataset have to establish a common understanding of the value of geographic information acquired from the dataset.
The main research question of this thesis is therefore: Is it possible to construct a computational model for quantifying the economic value of geographic information in a decision making process?
We are interested in building a computational model that will improve our understanding of the economic value of geographic information. Kuipers defines a model to be “a (small) finite description of an infinitely complex reality, constructed for the purpose of answering particular questions (Kuipers 1994)”. Computational models allow us to use formal tools for the model construction and to verify the validity of the model by executing the formally specified programs.
Goal and hypothesis
The goal of this thesis is to develop a computational model for assessing the value of geographic information for the user. With the construction of the computational, agent-based model we aim to improve our understanding of the issues of geographic data use and geographic datasets as an economic good, the value the users and potential buyers attach to the geographic information, and possible mechanisms of pricing, especially pricing according to the value geographic information has for a user. Our central hypothesis is that it is possible to construct an agent-based, formal model for quantifying the value of geographic information in relation to the decision making process in which it is used. The hypothesis positively answers the research question, and we assume that we can quantify the value of geographic information for a particular use.
We use a car navigation study case in order to answer our research question and verify the hypothesis. A simulated agent is navigating in the city of Vienna and its task is to find the shortest path from its current position in the street network to the preferred destination. The agent gets the information about the street network from the street network dataset. Navigation in a city represents a decision making process; every intersection is a decision making point where an agent has to decide which street to take to continue driving. The value of the geographic information in this decision making process is proportional to the time the agent can save by selecting the street to follow, compared to a random selection in the absence of this information.
The methodological approach for quantifying the value of geographic information in a decision making process used in this thesis consists of the following four parts:
· Value and price of an economic good
· Conceptual model for quantifying the value of geographic information
· Formalization of the conceptual model
· Testing the formal model
We explain the economic background for the conceptual model that is grounded in value and price theory and then we present the agent-based conceptual model. The third part consists of the formalization of the conceptual framework. Finally, we test the formal model by executing agent-based simulations applied to the case study of car navigation. The results of the simulations are crucial for verifying our hypothesis.
Frank, A. U. (1996). Der Nutzen und der Preis von Geographischer Information. AGIT'96, Salzburg, Proceedings of AGIT'96.
Krek, A. and A. U. Frank (1999). Pricing Geographic Data. GIM International, The Worldwide Magazine for Geomatics. 13.
Kuipers, B. (1994). Qualitative Reasoning, Modeling and Simulation with Incomplete Knowledge, The MIT Press.
Nelson, P. (1970). “Information and consumer behavior.” Journal of Political Economy(78): 311 - 29.
Shapiro, C. and H. R. Varian (1999). Information rules, A Strategic Guide to the Network Economy, Harvard Business School Press.
Download PhD-Thesis in PDF format