E X E C U T I V E S U M M A R Y
On the late morning of Friday the 13th August 1993, the most tragic collapse of building in the history of Thailand took place in Korat, Nakorn Ratchasima Province. The collapse took less than 10 seconds to reduce the 6-storey complex into a mammoth pile of rubble, burying 379 unsuspecting occupants, killing 137 and badly injuring 227. The sights of the rubble, the pitiful victims, and the strenuous rescue operation will remain forever in the memory of most Thai people of this generation.

After the incident, all parties concerned, both public and private sectors, expressed much interest in knowing the real cause, and in securing measures for future prevention. A technical investigation under the Civil Engineering Chapter of the Engineering Institute of Thailand (EIT) commenced immediately upon the arrival of the Task Force at the site on August 15, 1994. In addition to on-site inspection, technical data were collected and assessments made of the nature of the collapse. These information together with results of subsequent laboratory and model tests led to the conclusions appeared in this report.

Moment of The Collapse
At approximately 10:10 am. on August 13, 1993, the six-storey building of the Royal Plaza Hotel collapsed swiftly and totally, leaving high only the front elevator hall, which was built with structural separation from the rest. The collapse was mostly vertical and after the collapse, leaving a neat floor-to-floor stack of concrete slabs. For structural engineers, the impression of the collapse reminds of a professionally torn-down building with the use of explosives. The entire building moved downwards under its own weight within seconds and with no time to fall sideways.

At the time of collapse, the building was occupied by 379 people. Among them are 117 government officials from the Education Ministry were attending a seminar in Benchamas Room on the second floor, 59 from Shell (Thailand) in Pikul-Chuanchom Room on the fourth floor, and 78 as general hotel clients. The rest are 125 hotel staff. At the end of the rescue operation on September 3, 1993, it was reported 127 killed, and 227 injured

Historical Records
According to the construction permit granted in 1983, the original part of the building, used as massage parlor, was 3 storey high plus one underground level. Later in 1985, the local building authority approved a construction permit to modify the building for use as hotel. But not until 1990 that a permit was sought to add 3 more storeys to the original buildings (see Figure 2 in Report). Checking from the calculation sheets submitted by the structural engineer at the time of permit application, there was no record of foundation and column strengthening. There was no explanation how the modified structure could retain the same safety standard after the addition of 3 more storeys. Later at the request of the official, soil test was conducted by a local technical institution to certify that the safety margin of the existing foundation was sufficient in carrying 3 more storeys. No attempt was made further on the assessment of safety margin of existing columns. The permit was granted nevertheless.

Method of Investigation
The collapse in Korat was rather uncommon for the following reasons:

(a) Most building collapses occur either during or just after the construction. In this case, the collapse occurred about 3 years after the last modification.

(b) Most building collapses, after a long period of use, were caused by unexpected forces of nature, such as earthquake, gust, etc. At the time of collapse in Korat, however, there was no report of any harsh environment, nor report of any misuse of the building to an extent that can generate a devastating force of this scale that causes the swift and total collapse.

After the incident, various aspects of investigation were conducted by independent teams from Public Works Department (PWD), Asian Institute of Technology (AIT), Chulalongkorn University (CU), Kasetsart Unversity (KU), Khon Kaen University (KKU), and Engineering Institute of Thailand with members derived from most universities and private consultants. One conclusion agreed by all parties is that the structure of the Royal Plaza Hotel was substandard, i.e., having marginal factor of safety, after the addition of 3 more storeys.

The addition of 3 floor areas of dead and live loads over the same set of columns that were originally designed to carry only 3 floors means that the original safety margin could be reduced approximately close to zero.

Normally safety of a building is generously provided in the process of structural design. Because of the high factor of safety practiced by structural engineers, even if one part of a properly designed building is substantially overloaded and fails, redistribution of the loads will take place. In such cases only local failures, but not a total collapse, will result.

The case of the Royal Plaza Hotel was extraordinary as all columns were nearly at their respective limit. This defies the belief that if a building has survived past the construction stage, it is very unlikely that it can collapse abruptly without major external forces, such as an earthquake.

Judging from the mode of failure, it can only be concluded that the collapse started when all columns on the ground level failed nearly about the same time, causing the entire building to fall under its own weight. All these columns were overloaded and remained with only marginal factor of safety. A failure of one column could easily trigger the rest as the load carried by this column would be redistributed to the rest. This subsequently caused them to fail, one after another, within a very short time. This process is well known as "progressive failure". Within a few seconds, all columns on the ground floor collapsed, and the 7000 ton mass of the central part of the building was unsupported. AT this instance, it was like lifting the entire building 5 meters high and letting it to drop under its own weight. The great impact caused a compressive wave to the upper columns causing all of them to collapse shortly. From observation during the removal of the debris, all columns at the ground level burst almost into loose particles, a failure rather common in the laboratory test of concrete cylinders. However columns from the second storey up appeared to have buckled due to compressive wave after the great impact. In addition, most of the survivors told of having heard two distinctive loud noises seconds before the collapse. Presumably, the first loud noise is due to the bursting of all ground floor columns, and the next one due to the impact of upper-floor mass to the ground.

The First Failure.
In the professional point of view, no matter what triggered the first column failure, the main reason for the collapse is the marginal factor of safety after 3 more storeys were added to the building. Theoretically, it is still interesting to find out the possible causes. To be able to point them out, the Task Force of EIT have conducted the following tests:

(a) Test of Concrete Strength. On-site test of concrete strength using the non-destructive Ultrasonic Pulse Velocity Test (PUNDIT) revealed that the tested concrete samples had a compressive strength in the range of 10-15 MPa with 95 % reliability. This result did not, therefore, point to any deficiency in the quality of existing concrete. This was also confirmed by the laboratory test of concrete cylinders cored from the remaining parts.

(b) Test of Rebars. Reinforcing steel bars collected from the site were tested in the laboratory, and all were found to conform to standard.

(c) Soil Test. The laboratory test of boring log from a location in the vicinity of the collapse indicated that the bearing capacity of soil in the area was sufficient for the foundation even after the modification. Settlement based on the loading of 6 storeys is neglible. Settlement due to consolidation or caused by subgrade water is unlikely, judging from the composition of soil in the area. This conclusion coincided with that of the investigation team from Public Works Department after the debris over the area suspected to be the starting point was removed. After excavation, foundation was found to be in good condition.

After eliminate other unlikely causes, the most likely cause of column failure is due to creep in all ground floor columns under a high stress level. After adding 3 more storeys, some columns on the ground level carried a dead load up to 70-80 % of strength capacity. Under this sustained overloading, micro-cracks developed in the concrete and, over the years, they grew and propagated. The resulting creep deformation in these columns increase gradually. As these columns crept and became shorter than the neighboring columns, load was then redistributed to others, causing other columns to be more overload and creep too. In this manner, loads were redistributed back and forth over the period of 3 years until the structural deterioration gradually spread over all columns on the ground floor. When load in one column exceeded its capacity and collapsed first, a progressive collapse of all columns on the ground floor followed and completed within seconds. [1]

Lessons
Under the on-going economic fast-track experienced in Thailand, economic consideration seems to be the main reason that can dictate and deviate standard practices of many professions. While entrepreneurs want to cash in with minimum investment efforts, they should be far-sighted enough to invest a little more if this means a premium to prevent unacceptable risks. Most important step for them is, therefore, to be able to appreciate the independent advice of the right professionals. As far as the safety of buildings is concerned, the best bet against this type of man-made disaster is to acquire the service of an able and honest professional engineers, who care more for the standard of the works than to obey the architects and the clients.

Short-term Measures
Just after the incident, Prime Minister Chuan Leekpai visited the site and ordered public universities to help review the safety aspect of all buildings under suspicion throughout the country. All together 218 buildings were under investigation and a number of them received recommendations for renovation. This measure was able to reduce public concerns over the safety of existing buildings. In addition, a data base of buildings around the country can be established in the process.

Long-term Measures
Because of the Korat man-made disaster, all parties in the engineering profession expressed concerns over the current engineering practice. They seemed to agree that the profession needs a major re-structuring. Many meetings and seminars were organized afterwards with an attempt to find the long-term measures. Although the current status of actions is unclear to the Task Force at the time of preparing this report, the following four points of actions have been discussed widely:
(1) Registration of professional engineers, who are allowed by law to practice, must be more strictly controlled.
(2) If (1) is not effective, indemnity insurance system must be established.
(3) If (1) is not effective, design of all public buildings must be cross-checked and jointly responsible by an independent third-party engineer.
(4) Usage of public buildings requires a license which must be renewed every 5 or 10 years.

Establishment of Engineering Council
Currently, all building construction works are controlled by the Building Control Act 2522/2535 (1979/1992), and engineers must register to practice under the Engineering Profession Act 2505 (1962).

Under the Building Control Act, the authority who exercises the control of "works" is the Office of Building Control Committee, an affiliation with the Public Works Department. By the Engineering Profession Act, the authority who exercises the control of "professionals who conduct the works" is the Office of Engineering Profession Control Committee under Interior Ministry.

When a registered professional engineer submits engineering drawings in the application of construction permit, he must also accompany with a set of engineering calculations. It is stipulated in the Building Control Act 2522/2533 that local building authority need not verify these calculations.

The idea is to allow a registered professional engineer to take the full responsibility of his work. On the other hand, this also implies that a decision made by one single person may decide the fate of hundreds of lives in the building, a case of Royal Plaza Hotel disaster. The screening, qualifying and ethical control of professional engineers are therefore crucial in reducing the risk of similar incidents. This can be done effectively if the screening party also represent members of the same profession. A strong proposal is therefore the establishment of "Engineering Council" with a mechanism for its members to protect, control and look after the profession. This is also along the policy of the current government in bringing the private sector to support public interests. The same recommendation was also given in a report on "Strengthening Disaster Management Strategies in Thailand", a study by Asian Disaster Preparedness Center, Asian Insitute of Technology, for Royal Thai Government supported by the United Nations Development Programme.

Members of the EIT Task Force Task Force
Worsak Kanok-Nukulchai (Chairman)
Mhan Sriroenthong (Secretary)
Noppadol Phien-wej
Krai Tangsanga
Anek Siripanichkorn
Pennung Warnichai
Poonsak Piansusom
Boonchai Satitmanaitham
(Completion Date: September 30, 1994)

[1] Similar explanation was used for the cause of collapse of Hotel New World in Singapore on March 15, 1986. Chairman of the EIT Task Force wishes to thank Professor Sang-Lip Lee, who was a member of the Commission of Inquiry into the collapse of Hotel New World, for the opportunity to discuss this matter with him.