I have to solve the following problem:

I have a following problem with a translation service.

I have one person that is in charge of spreading the documents between the translators, so that all of them are under similar work load.

I need to create a productivity tool that will allow us to monitor the load. The input can be via some type of form, the output needs to be graphical (I was thinking in a form of chart), so that he can decide who gets what on the fly.

The person in charge of distribution will enter basic document properties of the document,

name,
length in pages,
deadline
name of the translator.

Upon receiving and reviewing the document, the translator will grade the document by difficulty of translation, lets say from 1 to 5. for example purposes

Normal load of complete pages/day is lets say 5. for every grade of difficulty, it would add a certain percentage of time needed for the translation.

The coordinatro should be able to open the report via web and monitor the load ... data would refresh itself every time he makes a new entry.

This is the basic idea ... I'm somewhat familiar with databases, mostly used access to ease my own work, however I never had to make the database interact with a web user interface or make it friendly for other users, so there is a new challenge.

What is the good approach to solving that problem ... what software is best for me to start chewing on.
Public license preferably.

Thank you!

Public license; mySQL and PHP are always a good combination.

As for the db design; start by identifying your entities (translators, documents, etc).

Whilst you're doing this you're going to want to think about how these things relate. Your choices are 1-1, 1-to-many and many-to-many (e.g. one translator can have many document assignments).

Once that is done you can start naming your entity's attributes (a translator has a name, a document has a number of pages, etc).

Give this a shot and post back your findings here for some comments

Do you want to store the documents within the database?

Will a document ever be split between two or more translators? (a large document, combined with a short leadtime, for instance?)

How many concurrent users do you anticipate?

georgev: mostly it is many to one, but sometimes as loquin said, it occurs that there is a large document distributed among several translators because of the short deadline.

at this stage Im not thinking about actual document distribution, more like the person that assigns everithing, to have a more clear picture of how much work load the translators are, so that the work is more or less equaly distributed.

So it is one or two persons at the time accessing the database at this stage.

With one or two people, then just about any database would support your needs. If you feel that it might grow in the future (from a concurrent user viewpoint) it may impact your current decision as the choice of database.

Will the person who decides on the document distribution ever translate any himself? If yes, they would have two roles. Is it possible for the role of assigning documents to translator ever be 'shifted' from one person to another, even temporarily (vacations, sick leave, etc)

Going back to georgev's post, what are your entities? Keep in mind that, essentially, each entity gets a table. What properties are associated with each entity? (For instance, translators would have a name, phone number, address, possibly an assigned skill level? Documents could have a name, path, file size, diifficulty level, ...)

Then, look at the relationships. How are the translators related to the documents. You've said that each translator can be assigned to many documents, and each document can be assigned to many translators. This is the definition of a many-to-many relationship, and it is modeled using an intersection table.

In any event, don't get the cart ahead of the horse. Document the entities, their properties, and their relationships. Take a look at the very first post in the Database Concepts & Design forum.

Another consideration, will you want translated documents reread / doublechecked. Sometimes one translator will use a particular turn of phrase that is not entirely apt to the context of the entire document, having a proof-reader might be a wise move. However you will have to allow for proof-reading time in your overall estimates of the duration of a translation.

How to determine the productivity in this scenario, any suggestions, please?

Workload can be calculated as perun said, by always assign a number of days to a document, number of das that is to be determined based on the difficulty of the document and some algorithm.

Therefore I assume that the distributor will give a document to be translated to the one that has the least amount of hours to be spent on his current documents.

Now there are other questions to be asked. Is a more difficult document supposed to be translated in more time, than a more easy one? or there is a deadline and the translator has to be more efficient?
Based on this you can talk about productivity, otherwise I suppose the productivity it's just a sum of the hours you spent translating documents per week/month.