`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`As described in the following claim chart, claims 1-12 of U.S. Patent No. 6,968,248 (“the ’248 Patent”) are invalid under 35 U.S.C.
`§§ 102 and/or 103 in view of U.S. Patent No. 5,260,868 (“Gupta”).1, 2 To the extent that Gupta is found not to anticipate one or more
`of the claims of the ’248 Patent, those claims are obvious in view of Gupta, alone or in combination with other prior art references,
`including, without limitation, one or more references identified in Exhibit D to Defendant’s Preliminary Invalidity Contentions.
`Defendant’s Preliminary Invalidity Contentions provide additional details regarding the motivation to combine Gupta and the
`references cited in those exhibits.
`Citations to particular documents or passages are merely exemplary of where each limitation is found. Defendant reserves the right to
`rely on other documents or passages providing comparable evidence of how Gupta alone or in combination with other prior art renders
`the ’248 Patent invalid.
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`1[pre] A method for scheduling in an automated
`manufacturing environment, comprising:
`
`Defendant does not concede that the preamble is limiting. To the extent
`it is limiting, see, e.g.:
`
`Gupta at Abstract: “A mechanism and method for calendaring a plurality
`of events such as scheduling the operation of interrelated machines
`which perform a process flow. Future time is divided into segments,
`called buckets, of increasing length. The first two buckets are of the
`same size and each of the following buckets twice as large as its
`
`1 U.S. Patent No. 5,260,868 issued on November 9, 1993 from an application filed on October 15, 1991, and is prior art under at least
`35 U.S.C. § 102(b).
`2 The ’868 patent is a continuation of a divisional of U.S. Patent No. 4,888,692 to Gupta. The ’692 patent issued on December 19,
`1989 from an application filed on November 10, 1988 and is also prior art under at least 35 U.S.C. § 102(b). Although only citations
`to the ’868 patent are provided, the disclosures identified herein are equally available in the ’692 publication. Defendant reserves the
`right to also rely on the ’ 692 publication as prior art in this proceeding for the reasons identified herein.
`
`- 1 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`preceding bucket. The first bucket slides so as to always cover a
`specified length of time following the current time. Events scheduled in
`the calendar is added to the appropriate bucket, depending on how far in
`the future it is to take place. When the current time equals the scheduled
`time for an event, then that event is removed from the bucket where it
`resides. When a bucket has become empty because all events have been
`removed from it, the events in the following bucket are distributed over
`the two buckets preceding it.”
`
`Gupta at 1:21-33: “The present invention relates to automated
`scheduling and planning systems.
`
`Resource planning is used extensively by industry. It is especially useful
`in the manufacturing sector, where careful scheduling of a
`manufacturing facility is necessary in order for such plants to be
`efficient. The flow of raw and partially finished goods, and scheduling
`of work on the various available machines, is a significant problem in
`large manufacturing facilities. A few examples of manufacturing
`facilities which are especially sensitive to scheduling problems include
`semiconductor fabrication facilities (front-ends), job shops, and plants
`making automobiles and heavy machinery.”
`
`Gupta at 2:52-66: “Therefore, a scheduling system includes a global,
`steady-state model of the entire manufacturing process. This global
`calculation is done one time and recalculated only when there is a major
`change in process flow definition or machine availability. This global
`
`- 2 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`plan generates parameters which are used to control local decision
`making strategies. The local strategies are applied to each machine in the
`manufacturing facility, and are relatively simple. Based upon the
`parameters extracted from the global definition, and information
`regarding the current state of the neighborhood of the particular
`machine, local decisions can be made on a real time basis. Special
`decision making strategies may be used by machines which are
`indentified as critical to the manufacturing process flow.”
`
`Gupta at 3:39-50: “The following description of the preferred
`embodiment includes detailed examples as well as the general
`approaches used in making a scheduling system. The description is
`broken into 4 major areas: a general description of a factory system,
`including definitions of terms found elsewhere; the global (steady-state)
`planning process: local planning and optimization; and a preferred
`calendar mechanism for use by the scheduler. It is understood that
`particular references and descriptions are not intended to limit the scope
`of the claims to the details shown therein, but are for illustrative
`purposes.”
`
`Gupta at 3:61-63: “The preferred scheduling system will be described
`with relation to a front-end manufacturing facility for integrated
`circuits.”
`
`Gupta at 4:1-10: “The scheduling system will be described with respect
`to a front end which is highly automated…. As will be described most of
`
`- 3 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`1[a]
`
`automatically detecting an occurrence of a
`predetermined event in an integrated,
`automated process flow;
`
`the control functions will be handled directly by the scheduling system,
`but it is a straightforward task to have some of these functions handled
`by the ma(cid:173)chines themselves if they are capable of doing so.”
`
`See also Gupta at FIGs. 2, 3, 5-7; Gupta at claim 1.
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`See, e.g.:
`
`Gupta at 4:44-47: “The process steps can be visualized as a long string
`of events which operate to transform a bare silicon slice at the first
`process step to finished integrated circuits at the last process step.”
`
`Gupta at 5:8-12: “If this process is a rework decision point, that is, a
`check or inspection process that might cause slices to branch into a
`rework loop as described above, a pointer to the start of the rework loop
`is kept.”
`
`Gupta at 6:9-13: “Each machine also has information showing its
`scheduled downtime. This includes both the frequency and expected
`
`- 4 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`length of such downtimes. Scheduled downtimes are those required for
`preventive maintenance, plant shutdowns, and other predictable events.”
`
`Gupta at 7:20-27: “The remaining items in the data structure of FIG. 3
`are related to the dynamic operation of the scheduler, rather than the
`steady-state structure of the machine as do the above described data
`items. The information concerning lots done on the current process and
`side are used in the local decision making process, or local optimization,
`of the machines as will be described under that section.”
`
`Gupta at 13:63-14:6 : “Once the global system parameters have been
`deter(cid:173)mined, each machine has several data structures which determine
`its behavior during operation of the manufac(cid:173)turing facility. These data
`structures act as a set of guideline instructions which tell each machine
`what to do next. Decision-making is event driven, and a determination
`of what comes next for each machine is made whenever certain events
`take place. Events which drive the decision making process include
`machine loads and unloads, and a machine going off-line or coming on-
`line. Whenever one of these events occurs, the scheduling system must
`calculate what that machine will do next.”
`
`Gupta at FIG. 7.
`
`- 5 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`See also Gupta at 7:27-61, claim 1; FIG. 3.
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`
`- 6 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`1[b]
`
`automatically notifying a software scheduling
`agent of the occurrence; and
`
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`See, e.g.:
`
`Gupta at 13:45-54: “A mathematical function is generated for each
`machine in the plant which incorporates the relevent factors, and global
`plans are made to minimize or maximize that function, whichever is
`appropriate.
`
`The information from the global planning stage is used to control the
`local decision making process. Each machine has a profile which
`indicates its place in the overall scheme; it will then take real time local
`knowl(cid:173)edge and combine it with this information to do local planning, as
`will be described below.”
`
`Gupta at 13:63-14:6 : “Once the global system parameters have been
`deter(cid:173)mined, each machine has several data structures which determine
`its behavior during operation of the manufac(cid:173)turing facility. These data
`structures act as a set of guideline instructions which tell each machine
`what to do next. Decision-making is event driven, and a determination
`of what comes next for each machine is made whenever certain events
`take place. Events which drive the decision making process include
`machine loads and unloads, and a machine going off-line or coming on-
`
`- 7 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`line. Whenever one of these events occurs, the scheduling system must
`calculate what that machine will do next.”
`
`Gupta at 14:20-25: “In the preferred embodiment, a single processing
`system runs the entire scheduling system. Since decisions are made on a
`local basis, a single moderately powerful processor can easily handle all
`the computational demands of a large, complex manufacturing facility.”
`
`Gupta at 27:58-68: “In order to implement the above described calendar
`mechanism efficiently in a computer, several preferred data structures
`are used. All buckets except bucket 0 consist of linked lists of events.
`Each event has an description identifying what is to occur, a time, and a
`pointer to the next event in the list. The time is a binary absolute number
`dating from the beginning of the use of the calendar. If 24 bit numbers
`are used, 16 million time steps can be calendared, which is equal to
`approximately 190 years. Larger binary numbers can be used if longer
`time periods are desired.”
`
`See also Gupta at claim 1.
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`- 8 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`NO.
`
`1[c]
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`reactively scheduling an action from the
`software scheduling agent responsive to the
`detection of the predetermined event.
`
`See, e.g.:
`
`Gupta at 5:8-12: “If this process is a rework decision point, that is, a
`check or inspection process that might cause slices to branch into a
`rework loop as described above, a pointer to the start of the rework loop
`is kept.”
`
`Gupta at 7:20-27: “The remaining items in the data structure of FIG. 3
`are related to the dynamic operation of the scheduler, rather than the
`steady-state structure of the machine as do the above described data
`items. The information concerning lots done on the current process and
`side are used in the local decision making process, or local optimization,
`of the machines as will be described under that section.”
`
`Gupta at 13:57-62: “The real-time portion of the scheduling system
`de(cid:173)pends on local optimization to function efficiently. In(cid:173)stead of
`recalculating the complete global state for the system each time a
`decision must be made only the relevant local state is recalculated. This
`greatly de(cid:173)creases the processor load.”
`
`Gupta at 14:7-12: “The range of actions which can be taken is fairly
`limited. A given machine may need to load a lot immediately, and the
`lot may need to be taken from one of several input queues. A machine
`which processes multi(cid:173)ple lots may be required to wait for a full load, or
`proceed with a partial load.”
`
`- 9 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`2
`
`The method of claim 1, wherein automatically
`detecting the occurrence of the predetermined
`event includes detecting an unplanned event
`or an unexpected event.
`
`See also Gupta at 7:27-61, claim 1; FIG. 3.
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`See, e.g.:
`
`Gupta at 4:53-61: “The string of process steps is not always a single
`string of events occurring in a fixed order. It is sometimes necessary to
`rework some slices at various stages of the process. For example, if for
`some reason a photo(cid:173)resist patterning step did not occur properly, it is
`neces(cid:173)sary to remove all of the resist, clean the slice, reapply photoresist,
`and redo the patterning step. This is re(cid:173)ferred to as a rework loop, and,
`on a schematic diagram of the manufacturing process, appears as a small
`loop of process steps off to one side of the main process flow.”
`
`Gupta at 16:34-48: “Broken machines will tend to develop large queues
`until they are fixed, even if the average usage is low. It is somewhat
`inefficient for the processes preceding the broken machine to keep
`feeding lots into the queue if the machines used for those processes
`could be utilized for other processes. Thus, a broken machine or perhaps
`
`- 10 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`even one which has developed a large queue through natural fluctuations
`in the flow of material through the facility, can send a negative demand,
`or lack of demand, signal to the upstream processes. This signal will
`tend to cause the upstream multiprocess machines to prefer processes
`which lead elsewhere than to the broken machine. This alleviates
`somewhat the build up of queues in the facility, with the corresponding
`increase in average cycle time.”
`
`Gupta at FIG. 1.
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`- 11 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`3
`
`The method of claim 1, wherein automatically
`detecting the occurrence of the predetermined
`event includes detecting an occurrence of an
`appointment state change.
`
`See, e.g.:
`
`Gupta at 6:9-13: “Each machine also has information showing its
`scheduled downtime. This includes both the frequency and expected
`length of such downtimes. Scheduled downtimes are those required for
`preventive maintenance, plant shutdowns, and other predictable events.”
`
`Gupta at 13:63-14:6 : “Once the global system parameters have been
`deter(cid:173)mined, each machine has several data structures which determine
`its behavior during operation of the manufac(cid:173)turing facility. These data
`structures act as a set of guideline instructions which tell each machine
`what to do next. Decision-making is event driven, and a determination
`of what comes next for each machine is made whenever certain events
`take place. Events which drive the decision making process include
`machine loads and unloads, and a machine going off-line or coming on-
`line. Whenever one of these events occurs, the scheduling system must
`calculate what that machine will do next.”
`
`Gupta at 16:34-48: “Broken machines will tend to develop large queues
`until they are fixed, even if the average usage is low. It is somewhat
`inefficient for the processes preceding the broken machine to keep
`feeding lots into the queue if the machines used for those processes
`could be utilized for other processes. Thus, a broken machine or perhaps
`even one which has developed a large queue through natural fluctuations
`
`- 12 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`in the flow of material through the facility, can send a negative demand,
`or lack of demand, signal to the upstream processes. This signal will
`tend to cause the upstream multiprocess machines to prefer processes
`which lead elsewhere than to the broken machine. This alleviates
`somewhat the build up of queues in the facility, with the corresponding
`increase in average cycle time.”
`
`Gupta at 19:40-51: “Assume further that the local prediction shows that
`the arrival time for the next lots into the queue for P20 is 45 time steps
`from now at which time 4 lots will arrive. It is easily seen that, including
`set up time two complete loads will be finished in 30 time steps, and a
`partial load could be finished within 40 time steps. Waiting for the
`additional lots to arrive will adversely impact the capacity of the plant. If
`possible, it is necessary to advance processing of the lots for P20 so that
`they will arrive by time 30 (from the current time). Ml accomplishes this
`by sending a demand signal to its up(cid:173)stream process. P19.”
`
`Gupta at 23:49-59: “Another type of control signal sent to upstream
`processes is a negative request, or lack of demand, signal. This signal is
`used when a machine is off line for repair or maintenance, and prevents
`large queues from build(cid:173)ing up in front of broken machines. Negative
`request signals also consist of placing a time that lots are needed in the
`data structure of the next upstream process. In fact, the negative request
`signal is the same as the normal request signal, except that the time step
`
`- 13 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`which is placed in the data structure for the upstream process is later
`than that for normal requests.”
`
`Gupta at 25:12-22: “Preferably, the negative request signal is only used
`for situations in which all machines available to a process are down for
`repair or maintenance. In some manufacturing facilities, it may be
`desirable to use a negative demand signal in other unusual
`circumstances. These could include any process which builds up a queue
`larger than some desired amount, or could be used when a near term
`plant shutdown is expected, and it is not desirable that certain machines
`have a queue during shutdown. In the latter case, the negative request
`will usually be a controlling signal.”
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`4
`
`The method of claim 3, wherein automatically
`detecting the appointment state change
`includes detecting at least one of an
`appointment cancellation, an appointment
`expansion, an appointment shrinking, an
`
`See, e.g.:
`
`Gupta at 8:48-60: “In the embodiment of the scheduler which is
`described herein, delays which occur between unloading a machine and
`making a lot available to the next process are not considered. Such
`
`- 14 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`appointment abort, an appointment changing
`status, an appointment shift, an appointment
`override, an transport time update, a load
`time update, an unload time update, a lot
`joining a batch, a lot leaving a batch,
`canceling a lot from a batch, and a
`commitment window update, leaving a batch,
`canceling a lot from a batch, and a
`commitment window update.
`
`delays are usually small com(cid:173)pared to the overall operation of the
`facility, and are not generally important. However, in cases where delays
`are significant, it may be necessary to take them into account. In such a
`situation, the transfer time is considered to be simply another process
`step, and is treated as are all other process steps. Thus, the overall
`scheduling system need not be modified to take such delays into
`account; they are handled within the parameters of the system as is
`currently described.”
`
`Gupta at 13:63-14:6 : “Once the global system parameters have been
`deter(cid:173)mined, each machine has several data structures which determine
`its behavior during operation of the manufac(cid:173)turing facility. These data
`structures act as a set of guideline instructions which tell each machine
`what to do next. Decision-making is event driven, and a determination
`of what comes next for each machine is made whenever certain events
`take place. Events which drive the decision making process include
`machine loads and unloads, and a machine going off-line or coming on-
`line. Whenever one of these events occurs, the scheduling system must
`calculate what that machine will do next.”
`
`Gupta at FIG. 7.
`
`- 15 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`- 16 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`5
`
`The method of claim 1, wherein automatically
`detecting the predetermined event includes
`detecting at least one of an appointment
`cancellation, an appointment expansion, an
`appointment shrinking, an appointment abort,
`an appointment becoming active, an
`appointment nearing completion, an
`appointment completing, an appointment
`shift, an appointment override, and a
`commitment window update, detection of a
`downtime occurrence; a machine becoming
`available; a PM/Qual being detected; a
`chamber going down; a chamber becoming
`available, a change in machine capabilities; a
`change in machine types; an addition of a
`process; an addition of a process operation; a
`lot arriving at a machine; a lot process
`changed, a lot placed on hold, a lot released
`from hold, a lot priority changed, a lot due
`date changed, a lot wafer count changed, a lot
`process operation changed, a lot departing a
`machine, an alarm firing for an appointment
`
`See, e.g.:
`
`Gupta at 6:9-13: “Each machine also has information showing its
`scheduled downtime. This includes both the frequency and expected
`length of such downtimes. Scheduled downtimes are those required for
`preventive maintenance, plant shutdowns, and other predictable events.”
`
`Gupta at 8:48-60: “In the embodiment of the scheduler which is
`described herein, delays which occur between unloading a machine and
`making a lot available to the next process are not considered. Such
`delays are usually small com(cid:173)pared to the overall operation of the
`facility, and are not generally important. However, in cases where delays
`are significant, it may be necessary to take them into account. In such a
`situation, the transfer time is considered to be simply another process
`step, and is treated as are all other process steps. Thus, the overall
`scheduling system need not be modified to take such delays into
`account; they are handled within the parameters of the system as is
`currently described.”
`
`Gupta at 13:63-14:6 : “Once the global system parameters have been
`deter(cid:173)mined, each machine has several data structures which determine
`its behavior during operation of the manufac(cid:173)turing facility. These data
`structures act as a set of guideline instructions which tell each machine
`what to do next. Decision-making is event driven, and a deter mination
`
`- 17 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`start time and an alarm firing for an
`appointment end time.
`
`of what comes next for each machine is made whenever certain events
`take place. Events which drive the decision making process include
`machine loads and unloads, and a machine going off-line or coming on-
`line. Whenever one of these events occurs, the scheduling system must
`calculate what that machine will do next.”
`
`Gupta at FIG. 7.
`
`- 18 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`Furthermore, this limitation would have been obvious in light of Gupta
`alone or in combination with references identified in Exhibit D relating
`to this limitation. Defendant’s Preliminary Infringement Contentions
`describes the motivations to combine Gupta and those references.
`
`- 19 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`6
`
`The method of claim 1, wherein automatically
`notifying the software scheduling agent of the
`occurrence includes:
`
`sending an indication of the occurrence to a
`publisher;
`
`publishing the occurrence from the publisher
`to a subscribing listener; and
`
`calling the software scheduling agent from the
`subscribing listener.
`
`See, e.g.:
`
`Gupta at 29:59-68: “A system for scheduling a semiconductor front end
`has been implemented consistent with the above description. It is written
`in Common LISP, and runs on an EXPLORER symbolic computer from
`Texas Instruments. A detailed simulation of a complete front end has
`been run, and the scheduler has proven capable of scheduling the factory
`at a speed greater than 1000 times faster than real time. This allows an
`entire month of scheduling, and simulation of plant operation, to be run
`in less than one hour.”
`
`Furthermore, claim 6 would have been obvious in light of Gupta alone
`or in combination with references identified in Exhibit D relating to the
`limitations of claim 6. This includes admitted prior art from the patent
`itself. See, e.g., ’248 patent at 8:1-11 (“More particularly, when the
`software agents 265 are created, they create listeners and subscribe to
`published events by adding the listeners to the event publisher. Listeners
`enable the software agents 265 to react to events in the process flow 100
`in an appropriate manner. Table 1, below, lists the relevant software
`agents 265 employed in the illustrated embodiment, the listeners they
`create, and a description of their function. Note that the use of publishers
`and subscribers via listeners and notifiers in this manner is known to the
`art, and any suitable technique may be employed.”). Defendant’s
`
`- 20 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`7
`
`The method of claim 1, wherein reactively
`scheduling the action includes at least one of
`aborting a scheduled appointment in
`progress; canceling a scheduled appointment
`duration of a scheduled appointment;
`shrinking the duration of a scheduled
`appointment; shifting a scheduled
`appointment; adding new processing
`capabilities; deleting old processing
`capabilities; setting an alarm; canceling an
`alarm; and changing the status of an
`appointment.
`
`Preliminary Infringement Contentions describes the motivations to
`combine Gupta and those references.
`
`See, e.g.:
`
`Gupta at 7:20-27: “The remaining items in the data structure of FIG. 3
`are related to the dynamic operation of the scheduler, rather than the
`steady-state structure of the machine as do the above described data
`items. The information concerning lots done on the current process and
`side are used in the local decision making process, or local optimization,
`of the machines as will be described under that section.”
`
`Gupta at 16:34-48: “Broken machines will tend to develop large queues
`until they are fixed, even if the average usage is low. It is somewhat
`inefficient for the processes preceding the broken machine to keep
`feeding lots into the queue if the machines used for those processes
`could be utilized for other processes. Thus, a broken machine or perhaps
`even one which has developed a large queue through natural fluctuations
`in the flow of material through the facility, can send a negative demand,
`or lack of demand, signal to the upstream processes. This signal will
`tend to cause the upstream multiprocess machines to prefer processes
`which lead elsewhere than to the broken machine. This alleviates
`
`- 21 -
`
`IPR2021-01342
`Ocean Semiconductor Exhibit 2027
`
`
`
`Exhibit D-4 to Defendant’s Invalidity Contentions:
`Comparison of U.S. Patent No. 6,968,248 and Gupta
`
`NO.
`
`’248 CLAIM ELEMENT
`
`GUPTA
`
`somewhat the build up of queues in the facility, with the corresponding
`increase in average cycle time.”
`
`Gupta at 19:40-51: “Assume further that the local prediction shows that
`the arrival time for the next lots into the queue for P20 is 45 time steps
`from now at which time 4 lots will arrive. It is easily seen that, including
`set up time two complete loads will be finished in 30 time steps, and a
`partial load could be finished within 40 time steps. Waiting for the
`additional lots to arrive will adversely impact the capacity of the plant. If
`possible, it is necessary to advance processing of the lots for P20 so that
`they will arrive by time 30 (from the current time). Ml accomplishes this
`by sending a demand signal to its up(cid:173)stream process. P19.”
`
`Gupta at 25:12-22: “Preferably, the negative request signal is only used
`for situations in which all machines available to a process are down for
`repair or maintenance. In some manufacturing facilities, it may be
`desirable to use a negative demand signal in other unusual
`circumstances. These could include any