This paper presents the optimization of multi-period bilevel supply chain planning for single supplier and single retailer with leader-follower relationship under demand uncertainty. The multiperiod production planning problem for the supplier and the retailer is formulated as the mixed integer nonlinear bilevel programming problem. In order to improve the total profit, the effects of the replacement of leader-follower relationship are investigated. The effectiveness of the replacement of leader-follower relationship is examined by some numerical experiments. A quantity discount contract with quadratic function is also proposed to maximize the total profit under demand uncertainty.
A new method of the process planning for end-milling operation considering product design constraints in this study. In our previous study, the process planning system, in which the Total Removal Volume is divided by the planes parallel with the XY, YZ or ZX planes to analyze machining sequence from top to bottom of the target product, is proposed. In this study, the process planning system, in which the Total Removal Volume is divided by all planes (including slope planes) existing on the target product, is proposed. Furthermore, the product design constraints or the designer's intention such as separate through holes which have the same central axis to be slide bearing is considered. A case study was conducted and the result showed that the proposed method can generate efficient multiple process plans for the machining operation. These multiple process plans or machining sequences are available to select adaptively the most suitable process plan or machining sequence under the several conditions such as the machine tool to be used and the product design constructions.
In this paper, we reveal an advantageous dynamic feature of singular conguration in dragging heavy objects with a two-link mobile manipulator. Unlike humans, the mobile manipulator can pull or drag the objects passing through its singular configuration where the manipulator is stretched out. The motion through the singular configuration can generate, by using the kinetic energy that is stored in the manipulator until the object starts to skid, much larger pulling force than the motion that avoids the configuration. Although the large force occurs only for a short period near the singular configuration, it can maximize the distance that the object is moved by the mobile manipulator under the condition that the consumption of joint torques is restricted. This dynamic advantage of singular configuration is verfied by numerical simulations and theoretical analysis.
Recently, while planning the design and operation of a production line, it has become common to simulate production conditions on a virtual factory. A virtual factory is constructed by virtual equipment. Virtual equipment is usually implemented in a virtual factory as a software agent which is a model of the real equipment. For a simulation of electric power consumption, behavior of the equipment should be modeled because power consumption depends on the equipment status such as operation and idling. Requirements for the equipment model were provided by the experimental implementation of a multi-agent based simulation system. In this paper, the method for developing an equipment model which fulfills the requirements is introduced. An equipment model has not only property descriptions but also behavior descriptions with computational expressions for calculating electric power consumption. Some of the variables in the computation expression are given values from the production schedule data, operation data and product data when a simulation is executed. Furthermore, to easily construct a virtual factory, the use of an equipment e-catalogue library is proposed. An equipment e-catalogue is generated by fulfilling values of the template. A template is a formal description of an equipment model.