What could AI Planning and Scheduling do for BPM?

After coming back from my presentation at BPM 2010 Demo Track (and my very first time in a conference of Business Process Management) about Smart Process Management, i would like to point out some thoughts I have in mind, mostly regarding at the future possibilities of using Artificial Intelligence Planning and Scheduling techniques in order to cover some of the gaps that still exists in BPM.

Even if using P&S for BPM can be very promising, I think that clearly the main research problem that have to be focused within this area is the dynamism that a Process can suffer during its life-cycle. When I say dynamism, I refer to the changes in the environment and/or the process model itself, that could happen once that the process has been started. These changes in the environment are the main reason for the emergence of a new area called “Adaptive Case Management”, and a very interesting panel discussion took place in the conference, driven by Keith Swenson, Dana Khoyi, Dermot McCauley and Jacob Ukelson. I was lucky to talk with Dermot McCauley, and what he told me is basically exposed in this excellent blog post at Column2, which briefly overview ACM as well.

In this panel discussion, K. Swenson gave a good definition of what ACM is, by using the “House” serial as example, and where decisions are made “on the fly” by the doctor, where the execution path cannot be predefined. Then, I came out with an issue: even if this is inevitable, doctors also try to do their own processes repeatable, capturing their models into what they call Clinical Guideline Protocols. So, as it is said in Column 2 blog, they look for a “happy path” showing what the patient treatment should be like. Hence, “this is not purely unstructured process, where there is no predefined model, but dynamic BPM where the model is predefined but can be readily changed while in flight.“. We can see that BPM and ACM can coexist.

So, if P&S techniques are going to be used in order to help to control and manage the technical problems associated with this dynamism, we must have clear in mind that P&S should be able to manage *at least*:

* exogenous events that can occur, even without having previously been foreseen in the process model, that changes either the model or the execution path.
* changes in the process model itself (skip tasks, add new tasks).
* changes in the enviroment (conditions of the knowledge workers, new knowledge workers, controlling the busy or available states).
* different (and possibly concurrent) execution paths, i.e. cases can be interrelated.

From my point of view, I see that P&S could be able to offer these features, by interleaving different planning stages, where the start and end points of these planning stages are delimited by the time at which the changes occur. Furthermore, there is already strong research open in continual planning, plan monitoring, re-planning and plan repair, which is similar and can be used for the problems arisen in this post.

Update: More insights into our very first approach in this direction can also be explored in the paper “Integrating plans into BPM technologies for Human-Centric Process Execution” presented at KEPS 2010.

an excellent AI human example

In complex problems, it is impossible to search the space exhaustively for the very best solution. Suppose, for example, you wear four different articles of clothing (i.e. shirt, trousers, socks and shoes) and you have 10 pieces of each article (10 shirts, 10 trousers, …). Then, there are 10.000 different combinations of clothes you can might wear each day, but no one has the time or interest to consider all the possibilities. Instead, people rely on heuristics, which are rules of thumb that contribute to satisfactory solutions without considering all the possibilities. A heuristic such as “wear brown shoes with brown trousers but not with black trousers” helps to provide an efficient solution to the problem of planning what to wear. Problem solving, learning, and language use can all be described in terms of rule-based heuristic search through a complex space of possibilities.

Source: Paul Thagard. “MIND: Introduction to Cognitive Science.” Second Edition, MIT Press, Cambridge, Massachussets.

Próximo destino: New York

Después de cruzar el charco en 2005 para ir a Perú, lo vuelvo a hacer, pero esta vez para acercarme a la Big Apple. Manhattan me espera para pernoctar, y más concretamente New Jersey, al otro lado del Río Hudson, será mi destino, para asistir en esta ocasión a una demo session en el congreso BPM 2010, donde el jueves 16 de Septiembre tendré que realizar una presentación, además de este póster.

La verdad es que me apasiona la idea de visitar por primera vez EEUU, justo en mi primer tercio de siglo, es un buen modo de celebrarlo. También me alucina poder presentar mi trabajo allí, por supuesto.

Manhattan skyline...

Por ahora me aseguro una visita al Top of the Rock para ver el skyline al atardecer, y otra visita al Radio City Music Hall, donde con suerte podré ver en directo al grupo de música Pop Vampire Weekend. Prefiero esta opción al tradicional musical en Brooklyn…qué le vamos a hacer, antes muerto que sencillo.

Se aceptan sugerencias, que se que los que me leéis sois mu viajeros. Otra cosa muy distinta es que me de tiempo… ojalá :)

WAI Meeting, 21 June

Rianne van Lambalgen : An Agent Model for Analysis of Human Performance Quality
A human’s performance in a complex task is highly dependent on the demands of the task, in the sense that highly demanding situations will often cause a degradation of performance. To maintain performance quality usually extra effort has to be contributed. However, the resources for such extra effort available to the human are limited. In this paper an agent model is proposed in which different types of relations between effort, task demands and performance quality can be used to analyse the human’s performance quality. It is illustrated how a support agent incorporating this model can support a human based on different performance criteria. The agent model thus allows to build agent applications that provide optimal support depending on a specific situation and goal of the task.

Anita de Waard : The Future of the Journal – A Proposal for Workflow-Based Science Publishing
We propose a conceptual format that forms the basis of a truly new way of
publishing science. In our proposal, all scientific communication objects
(including experimental workflows, direct results, email conversations, and
all drafted and published information artifacts) are labeled and stored in
a great, big, distributed data store (or many distributed data stores, that
are all connected). Each item has a set of metadata attached to it, which
includes (at least) the person and time it was created, the type of object
it is, and the status of the object including intellectual property rights
and ownership. Every researcher can (and must) deposit every knowledge item
that is produced in the lab into this repository. With this deposition
goes an essential metadata component that states who has the rights to see,
use, distribute, buy or sell this item.

There are two things needed to make this vision a reality: first, the
development of an exact, rich, future-proof set of metadata tags, which are
versatile enough to handle all the tasks described above, but not so
enormous that the system or the user are bogged down by them. Secondly,
tools need to be developed that allow the efficient storage, markup,
linking and retrieval of the multifarious data items that are to be added.
Both of these are close to being available.

WAI meeting, 7 Junio

Yan Wang : Improve Medical literature Search through Interest-based Query refinement
With the fast growing life science literatures on the Web, for scientific researchers, the difficulty of finding the most relevant results is especially obvious when they do not have enough experience to formulate their queries that exactly reflect their needs. While, the users’ previous knowledge background (more specifically, user interests) can be acquired through their publications. Hence we propose to provide user interests as contexts for literature search.

Michel Klein : Understanding behaviour change
Treatment for chronic diseases often consists of a combination of lifestyle advices and medication. We are developing an intelligent system with the aim of improving adherence to therapy; often, this implies supporting people in changing their behaviour. The basis of the system is a computational model of factors influencing behaviour. In the talk, I will present you the conceptual version of this model and ask you to participate in a small experiment to validate and calibrate the model.

Screencast/video demo for JABBAH

I did a (hopefully) nice screencast/video by using iMovie, which I have to say is really easy to use. The screencast is about our tool JABBAH and some experiments we did about an example process for “hospital patients admission”.

Feel free to comment about it, give any suggestion, critique, etc… we want to submit it to BPM 2010 Demo Track. Don’t forget to click on 720 HD for high definition!

WAI may 10th

Matthijs Pontier : An Affective Agent Playing Tic-Tac-Toe as Part of a Healing Environment

There is a growing belief that the environment plays an important role in the healing process of patients, supported by empirical findings. Previous research showed that psychological stress caused by loneliness can be reduced by artificial companions. As a pilot application for this purpose, this paper presents an affective agent playing tic-tac-toe with the user. Experimenting with a number of agents under different parameter settings shows the agent is able to show human-like emotional behavior, and can make decisions based on rationality as well as on affective influences. After discussing the application with clinical experts and making improvements where needed, the application can be tested in a clinical setting in future research.


Get every new post delivered to your Inbox.