The thesis describes the development of a new principle for flexible assembly cells, MARK III. The principle has been realised in a pilot installation in the laboratory at IVF-KTH Woxen Centre. The cell is a hybrid flexible assembly cell, which enables manual operations to be integrated at any stage in the assembly sequence.
MARK III is characterised by a reversed material flow, i. e. the robot is moved to the parts instead of the parts being fed to the robot. The reversed flow is made possible sinte the robot is mounted on a track motion. The assembly is carried out in fixtmes placed on an assembly-table mounted in hnt of the robot The sub-batch principle is used in order to reduce time losses due to gripper exchange and moving of the robot. All grippers are placed behiud the robot and exchange of grippers is done while travelling to the next feeding station.
The assembly fixtums are placed on a pallet and the pallet may be shifted to and korn a pallet magazine. The magazine is used both as a storage for non-used palleta and as a buffer between the automatic system and the manual assembly station. This solution de-couples, timewise, the operator from the automatic system thus providing a satisfactory work environment.
The MARK III-principle allows the parts to be fed by using any kind of feeding equipment. Robot vision is extensively used for parts feeding. The use of vision makes it possible to use low-cost feeding equipment, for example flat trays.
Seven requirements were formulated for the system principle:
Enable stepwise automation
Co-existencc of manual and automatic operations
Large product/variant flora
Enable use of many different feeding solutions
Large capacity span
Possibilities to integrate non-assembly operations
Easy and low-cost programming
The evaluation clearly states that the system principle fulfils these requirements. Even in the pilot installation the requirements are almost fullfilled.
A cell control system talled FACE (Flexible Assembly Control Euvironment) was developed at KTH in parallel with the development of h4ARK III. FACE is based on a relational data-base tun on a PC. The FACE system reduces the programming time and makes the system easy to run by un experienccd operators.
Mark III has been developed to be flexible enough to cape with assembly of many variants, in small batches and small ammal numbers. The flexibility is also making the system general, it can be adapted to many applications. MARK III can therefore be a standardised assembly cell and many systems may be installed in industry. This will cause a cost reduction due to the series effect.
For tbc evaluation of the system principle, five possible industrial applications have been studied Three of these were carried out in parallel and show several similaritics. Some components are the same in these systems; robot, assembly-table, pallet magazine, feeding principles and manual assembly station.
These systems show that MARK III can be a standardised assembly cell, which can be adapted to several applications. The studied products to be assembled are; universal joints, gear shafts. electrical motors, contactors and connectors.
An evaluation from an economical aspect has been made. It indicates that MARK III systems have a potential to give benefits to the user, i.e. the user must work actively in order to take advantage of these benefits. If the user only uses wage reduction to justify the investment, tbc system must be run in at kast two shifts.
MARK III systems are suitable for annual production numbers from approx. 400 000 to approx. 400 000 products per year. The necessaty armttal number is affected bom by the size of the product and by the equipment used in the system.
Institutionen för produktionssystem , 1998. , 131 p.