linjc16 / CT-FP

Demos for "Failure Response to Process Module in Cluster Tools with Residency Time Constraints Based on Two-Stage Mixed Integer Linear Programming".

Failure Response to Process Module in Cluster Tools with Residency Time Constraints Based on Two-Stage Mixed Integer Linear Programming

This repository provides some demos for Failure Response to Process Module in Cluster Tools with Residency Time Constraints Based on Two-Stage Mixed Integer Linear Programming.

Requirements and Installation

• Matlab 2018b

• YALMIP

• Gurobi Optimization

Getting Started

We release 3 files, named cyclic_demo.m, FR_first_stage_demo.m and FR_second_stage_demo.m, which represent MIP demos for cyclic scheduling MIP model, first-stage IP model and second-stage MILP model. We will give more details of each file in the following. You can just change the input to run your own experiments.

Cyclic Scheduling

The meaning of each input in cyclic_demo.m is as the following:

• robot_mode: single or dual

• n: the number of PMs in the CT

• w: the robot loading or unloading time

• v: the robot rotating time

• r: $r_i$ denotes the processing time in ${\rm PM}_i$

• residency: the $i$-th element denotes the longest residency time allowed in ${\rm PM}_i$

• K_PM: wafer flow pattern (WFP) in the original paper

Failure Response

First-stage Model

For the input in FR_first_stage_demo.m:

• robot_mode: single or dual

• w: the robot loading or unloading time

• v: the robot rotating time

• WFP: wafer flow pattern in the original paper

• MT0_PM: initial PM state

• MTn_PM: target PM state

Second-stage Model

For the input in FR_second_stage_demo.m:

• robot_mode: single or dual
• w: the robot loading or unloading time
• v: the robot rotating time
• WFP: wafer flow pattern in the original paper
• rho_raw: the $i$-th element denotes the processing time in ${\rm PM}_i$
• rho: eliminate elements in rho_raw that correspond to PMs that have failed.
• varthetaI: initial processing time state
• varthetaE: target processing time state
• MT0_PM: initial PM state
• MTn_PM: target PM state
• varpi_0: initial robot state
• varpi_1: target robot state
• delta: longest wafer residency time vector
• A_PM_idx: used for extracting $A_{\rm PM}$ from matrix $A$
• A_R_idx: used for extracting A_{\rm R} from matrix $A$
• Zl_idx: used for extracting loading action from Z
• Zu_idx: used for extracting unloading action from Z
• step: number of tasks that the robot will perform $k$

Contact

Please feel free to submit a Github issue if you have any questions or find any bugs. We do not guarantee any support, but will do our best if we can help.