Commit 800b3de0 authored by Markus Klinik's avatar Markus Klinik

how does our method fare in OPD?

parent 9c6236ff
......@@ -7,7 +7,7 @@ It is a variant of the MSRCPSP, and extends it with resource affinity constraint
We use an evolutionary algorithm to search the solution space for schedules that have a good score according to the weighted product of all objectives.
We argue that evolutionary algorithms lend itself well for our problem, because we are not only interested in the best schedule, but also in alternative solutions.
Evolutionary algorithms calculate sets of solutions, which as a by-product contains the best alternative solutions it has found.
Evolutionary algorithms calculate sets of solutions, which as a by-product contain the best alternative solutions it has found.
Furthermore, evolutionary algorithms do not need knowledge about the function they are trying to optimize.
This is useful for our purpose, because user-defined quality functions can be arbitrary Turing-complete calculations, and there can be no general heuristic that could take advantage of them to guide the search.
......@@ -16,23 +16,33 @@ We envision our algorithm to be part of an integrated mission management system,
Humans and machines should contribute to the joint activity of mission management with their respective strengths.
In his Ph.D. thesis, \citet{Johnson2014} argues that for interdependent teamwork, the computer must not be a black box.
Its internal process must be exposed.
In order to trust and rely on the computer, it must have three key properties: \emph{observability}, \emph{predictability}, and \emph{directability}.
He defines them as follows.
Observability is making relevant aspects of one the machine's status, knowledge and environment available to others.
In order to trust and rely on the computer, it must have three key characteristics: \emph{observability}, \emph{predictability}, and \emph{directability}.
Johnson defines them as follows.
Observability is making relevant aspects of the machine's status, knowledge and environment available to others.
Predictability means that others can rely on their prediction about the machine's behaviour.
Directability is the ability to be influenced by others.
This can be for example through direct commands, guidance, preferences, or suggestions.
We now look at our method with respect to these three characteristics.
Observability in our case means that the user can see what the machine knows about the ship's status, in other words the user should have access to the operational picture.
If the user defined quality functions make use for example of the weather conditions, the location of people on board, or the degradation status of equipment, this information should be accessible in a clear and user-friendly manner.
Representation and display of the operational picture is out of the scope of this article, so this point is not applicable to the discussion.
Predictability in our case means that given similar situations, the algorithm should come up with similar solutions.
The probabilistic nature of evolutionary algorithms makes our method come off less well in this respect.
For some problem instances, where good solutions are isolated between many worse solutions, we observed that it can take a long time or lots of runs to find a good solution again that we have seen before.
We believe that this problem can be mitigated by developing parameter sets that are appropriately dimensioned for typical problems.
Directability in our case means that the user can guide the search towards solutions that fulfil certain desired properties.
This is where the strength of our method lies.
User-defined quality functions and the weighted product were specifically included to give users lots of freedom to express preferences about the kind of solutions they desire.
The example in \cref{sec:example-conflicting-objectives} shows how emphasizing the makespan produces quick solutions where less capable resources are being utilized, while de-emphasizing the makespan assigns more work to fewer high-quality resources, resulting in schedules that take longer to complete.
The example in \cref{sec:example-search-and-rescue} demonstrates how arbitrary constraints can be encoded in user-defined quality functions, in this case to prevent Bob from having to fly the helicopter.
\begin{itemize}[noitemsep]
\item Why no machine learning AI?
\item Johnson, coactive design: computer must be no black box (Fong)!
\item OPD observability, predictability, directability
\item ML fails dramatically in all three disciplines
\item How does EMO perform in these disciplines?
\item The decision whether a boat or a helicopter should be used can be seen as a planning problem, not a scheduling problem.
\item Planning and scheduling overlap when a resource is a case.
\item Stop criterion: keep calculating until convergence instead of fixed number of rounds
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment