The internal target for certification of Comac’s 90-seat ARJ21 regional jet has been delayed again, to 2014 from the previously scheduled mid-2013, says a program official. Development therefore is now likely to last 12 years, with six years of flight testing. A key reason for the latest delay is slow progress in flight testing. Engineers also are attending to avionics bugs and modifying the landing gear, one of many issues arising from changes elsewhere in the design.

Another official working on the ARJ21 says that he knows of no formal decision to delay to 2014 but that, given the current state of development and flight tests, the aircraft cannot be certified before that year. It is quite likely to be ready in 2014, however, he adds.

While struggling to get the ARJ21 certified, Comac is considering both updating the aircraft, whose technology is aging even before airlines begin using it, and launching another regional airliner, say industry officials. In the end, the two ideas might be merged. The idea of improving the ARJ21, most obviously with new engines, has been looked at since about the time of the first flight in 2008.

The other project is sometimes called the New Regional Aircraft. It would follow the C929 widebody airliner that Comac proposes building once it completes development of the C919, a 158-seat narrowbody planned for certification in 2016 and intended to challenge the Airbus A320 family and the Boeing 737.

Among the ARJ21’s developmental problems is one emphasized by Comac internally: Coordination with the Civil Aviation Administration of China (CAAC) has not always been smooth. CAAC officials need to attend every test flight, but bringing together the schedules of the aircraft and the personnel from different organizations has not been easy and has delayed progress.

The CAAC also has been learning to improve its performance in certification programs, and not only in relation to the objective of ensuring safety. It has picked up experience in working more smoothly with Comac and suppliers. For example, it has not always been prompt in executing tasks such as inspecting materials and equipment, says a program official, adding that it can be expected to perform better in the future.

The growing experience of CAAC and Comac is one reason not to think that the ARJ21’s problems necessarily foreshadow similar delays in the much larger and more important C919 effort. In working on the C919, Comac has come to understand much more about commercial aircraft design and development than its predecessor, Avic Commercial Aircraft Co., knew during the early work on the ARJ21, the officials point out. The company knows more partly because of mistakes made in ARJ21 development.

Those errors have included a structural weakness—the ARJ21’s wing failed a strength test. One change leads to others, some at the insistence of the CAAC, such as modifications being made to the landing gear to meet revised load calculations. Sometimes the program has moved too slowly to address emerging issues, says an official.

“Some people say the CAAC is being tougher with Comac than the FAA would have been,” says that official. That should not be a surprise. The Chinese authority, renowned for its conservative approach to safety, is also unusually demanding in the standards it imposes on airframe maintenance, while its requirements for the physical condition of civil pilots are sometimes half jokingly compared with the national space program’s demands on astronauts.

The CAAC and the FAA are using the ARJ21 program as an exercise through which the Chinese authority will learn the ropes of certifying a commercial jet aircraft with Western agencies. Once the process is complete, CAAC certifications will be recognized by the FAA and will therefore be widely accepted in Western markets.

Full-scale development of the ARJ21 was approved in 2002 with the objective of putting the aircraft into service in 2006—and learning how to develop a commercial aircraft to Western standards. Managers soon realized that 2006 was too ambitious, and within weeks the target was slipped to 2007, with a first flight in 2005. When 2007 came around, the ARJ21 was due to go into service in 2009, but it did not fly until 2008, at which time the first delivery was pushed back until 2010. More delays followed. Last year, the target was September or October 2012, but early this year it was bumped to June 2013.

There may have been two bumps since then, as a fourth program official says the latest advice to Catic, the state enterprise responsible for international sales, is that late 2013 is the target. It seems that Catic has been told of one delay but not a second, to 2014.

As a Chinese state organization, Comac is not in the habit of announcing program delays, as a Western manufacturer would do. It also has good reason not to be quick in letting even its suppliers know that there has been another schedule slip, says an industry official.

“If suppliers think that they have more time, they won’t try so hard,” says that official. ARJ21 suppliers include Comac shareholder Avic (which builds the airframe), General Electric (CF34-10A engines), Honeywell (flight controls), Parker Aerospace (fuel system), Rockwell Collins (avionics) and Liebherr-Aerospace Lindenberg GmbH (landing gear). All are waiting to book volume-production revenue from their development investment.

Comac Chairman Jin Zhuanglong has criticized Catic for failing to secure many—perhaps any—dependable orders for ARJ21s from foreign customers. Catic’s response is that Comac, having no record as a supplier of serviceable airliners, needs to certify the ARJ21 first. Operations by the Comac-owned first customer, Chengdu Airlines, should help, too.

Some in the industry are concerned that Comac is losing interest in volume production of the ARJ21 and will be satisfied if it just achieves certification, supporting the C919. But the company has ordered parts for at least 13 aircraft, of which only the first four are prototypes not intended for airline service. It is therefore committed to building at least nine production aircraft.

The original ARJ21 program envisaged building 340 aircraft for sale to Chinese airlines. There were to be exports, too, but partners put much more faith in the domestic sales, which the Chinese government could force.

In assessing whether to use the ARJ21 as the basis for the New Regional Aircraft, a key issue is likely to be the size of the new model and the current airliner’s body cross-section, which accepts five-abreast economy seating and therefore could be stretched to fit 150 seats in two classes. For that reason, the ARJ21 more likely would be the basis for an aircraft that would compete against the Bombardier CSeries or the Sukhoi Superjet 100 than for a model with fewer than 100 seats.

Comac has said the cross-section, unusually wide for ARJ21’s 90 single-class seats (or 78 in two classes), was chosen for passenger comfort. But the real reason is that the ARJ21 is generally based on the five-abreast McDonnell Douglas MD-82, built at the same Shanghai factory in the 1980s and 1990s. The ARJ21 is not an exact copy, however.