SA_logo_text_small
SA_slogan1

>> Home >> Services >> Integrated Capacity Planning
 

Introduction to Integrated Capacity Planning

Integrated Capacity Planning (ICP) is the integration of Inspection Planning and regular fab capacity planning. ICP simultaneously plans the process, inspection, metrology, and AMHS* capacity. Just as in the Inspection Planning service, two important elements are included:

  1. The costs and benefits of inspection are used to calculate the inspection capacity.
  2. The impact that sampling plans have on process tool capacity, cycle-time, and fab throughput is accounted for.

The elements in ICP that differ from Inspection Planning are:

  1. The modeling of all process tools, not just the ones responsible for yield impacting defects/excursions.
  2. Capacity planning for metrology tools taking costs and benefits into account as far as data is available.
  3. Planning of AMHS capacity.
  4. Ability to study fab layouts. For example, the different impacts that farm and hybrid layouts have on excursion risk and yield in addition to overall tool budget.

*AMHS = Automated Material Handling System.
ICP_logoPicMedium

ICP, the full fab capacity planning service offered by Sensor Analytics

ICP integrates Inspection Planning with well established fab capacity planning methods. This service is also driven by our Inspection Planner software, IP1.0. This is possible since IP1.0 is by design a fab capacity planning model. The unique algorithms needed to study the costs and benefits of inspection & metrology are added on top.

Please refer to the software pages to learn about the standard capacity and queueing model capabilities of IP1.0. These features allow calculation of the tool capacity required to meet specific throughput and/or fab cycle-time targets.

 

The importance of integrated capacity planning

Sensor Analytics is the first to offer integrated capacity planning. Recent research results by Sensor Analytics have shown the importance of this new approach [see ISSM 2005 paper].

The main results are the fundamental connections between AMHSs, process tools, inspection tools, & the current sampling plan. These connections show that leaving the sampling plans as a static variable can lead to sub-optimal capacity and layout plans.Two examples of these connections are shown below.

Example 1: the yield benefit of AMHS systems

Sensor Analytics was the first to publish results on model based prediction of AMHS yield benefits.

In our research a detailed simulation model was used. We showed how the delivery speed of lots to inspection tools in a M1 layer affects M1 yield.

Our paper then extrapolates that a 37% reduction in lot transit time can result in approximately 4% fab wide yield increase. This is yield otherwise lost to excursions.

Sensor Analytics doesn’t sell AMHS or suggest that faster systems are always better. We are providing the capability to investigate and identify the economic options. This can be done with our IP1.0 software. IP1.0 captures the important dynamics of detailed simulation with fast analytic calculations.

Example 2: preventive maintenance, excursion detection, and cycle time

Published inspection planning models do not account correctly for the role of preventive maintenance (PM). Our software, IP1.0, does in two important ways:

  1. IP1.0 allows PM to “compete” with inspection tools, i.e. PM can fix excursions if it is done before inspection detects the excursion.
  2. IP1.0 captures the impact on fab cycle-time and throughput due to process tool PMs.

In our recent research we show how the queue time at a CMP tool group is impacted by different PM lot intervals. Since inspection tools are actively sampling to detect excursions, frequent PMs have a miniscule impact on the number of good wafers out, but a significant queue/cycle time impact.

Since overall project structure is identical, refer to the pages on Inspection Planning to learn about the specifics of an ICP project. Note that the relative ICP project scope and duration increases proportionately to the amount of process tool, metrology, and AMHS data that needs to be collected and pre-processed. Due to the larger scope, greater fab involvement is needed than for Inspection Planning.
ICP_fig2
ICP_fig1

Copyright 2006, Sensor Analytics.