Abstract:

IB (Insolvency and Bankruptcy) tasks are ill-known processes. Real life decision making related to IB is based on common sense reasoning only. To formalise such vague, sparse, partially inconsistent and subjective tasks as IB classical decision making trees can be used. However, traditional evaluation of these trees requires a complete set of relevant numerical values. These values are often either prohibitively inaccurate or they are unknown as they are unique. Moreover, decision accountants / economists dealing with IBs are usually not willing to invest too much time into study of complex mathematical theories. They require such decision algorithms which can be (re)checked by human like common sense reasoning. IB problem is transferred into a decision making tree which indicates all sub-decisions and random lotteries. Random lotteries quantifies such sub-IBs events which are not certain. Such decision trees are evaluated using well known methods. However, the required data sets are often incomplete. The problem of missing data items is solved by introducing a simple and generally applicable heuristic – a longer sequence of events is less probable. A company, which fell into bankruptcy hearing has legislatively supported several options how to deal with this situation. Each of the option can be solved in three different ways by using decision–making tree as tool for probability determination, using already known probabilities from previous case. A realistic case is presented in full details.