[en] In September 1999, in Tokai-mura during the termination operation of the JCO criticality accident, gamma and neutron radiation were measured. The doses of the workers in the vicinity of the sedimentation tank were about 50 times larger than pre-estimation. In this report, the reason why the doses of the workers were 50 times larger than pre-estimation was investigated and considerations for the cause was made. Furthermore, an appropriate assessment method of the radiation in the case of a similar accident was considered. In the criticality accident, the measurements by the Rem-counter were valid. However, there had been existed the scattering and shielding of gamma ray neutron by the building materials including cement and water etc. In particular, data of a distance of over 100 m should be carefully examined. So the pre-estimation dose was basically based on the data over 100 m distant point, the estimated value showed low dose. In this report, re-assessment of the dose based on the data from 40 m to 100 m range. As a result, as measured from 20 mSv/h to 3 mSv/h data are used which correspond to about three to four points. The re-assessment revealed that the assessment correspond to the measured dose within 60 to 80 percent accuracy. Under a high neutron dose field (the workers dose estimated around 100 mSv), in order to estimate the dose of the workers, (1) annual dose limit, (2) dose exposed from other work, (3) measurement error, have to be considered. Considering the above three points, 10 mSv (half of the annual dose limit 20 mSv) thought to be appropriate tentative dose level for the dose estimation itself. (author)