This is an appeal from the decision of the Patent and Trademark Office Board of Appeals (board) affirming the examiner’s final rejection under 35 USC 101 of method claims 1, 6, and 8, all the claims in appellant’s (Deutsch’s) application serial No. 86,-296, filed November 2, 1970, and entitled “Multi-Unit Optimization.” We reverse.
The Invention
Deutsch controls and optimizes the operation of a system of multi-unit plants — e. g., oil refineries at different geographic locations. The tools and techniques needed to achieve similar operation of a unit within a single plant were referred to by Deutsch as known in the prior art.1 Deutsch adapts that technology to the operation of a multiplant system. The claims on appeal are:
1. The method of operating a system of multi-plants which produce finished products from material derived from a plurality of sources at fluctuating costs for delivery to markets of variable prices, each of said plants having a different, unique, cost function in producing the products, said method comprising operating computing apparatus to automatically perform the steps of:
(a) establishing optimized control points for process units in a first plant,
(b) substantially continuously feeding material and energy cost data, product price data, and variable process data to the optimized controls for said process units to maintain them at operating conditions dependent upon said data,
(c) periodically modifying cost and process data fed to the control for said first plant to set operating points for said process units in said first plant dependent upon optimum operation of said first plant, and
(d) periodically modifying cost and price data fed to the control for said system to set an operating point for said plant dependent upon optimum operation of said system.
6. In the control of a system of multiunit plants which produce finished products from materials derived from a plurality of sources of supply wherein costs and market prices are subject to time variations and process variables are controllable, each of said plants operating at a different cost function, the method which comprises operating computing apparatus to automatically perform the steps of:
(a) closing an optimizing control loop on a process unit in a first plant,
(b) substantially continuously feeding material input functions, product output functions, and variable process functions to said control loop for maintaining said process unit at an optimum operating condition dependent upon said functions,
(c) periodically closing a control loop on said first plant to modify input and output functions fed to said control loop for said first plant periodically to establish an optimum plant operating point to set an operating point for said unit dependent upon optimum operation of said first plant, and
(d) periodically, but at less frequent intervals, closing an optimizing control loop *691on said system to modify cost and price functions fed to the control loop for said system to establish an optimum operating point and to reset operating points for said first plant for said process unit dependent upon optimum operation of said system.
8. In the control of a processing system including a plurality of plants which cooperate to produce finished products from materials derived from a plurality of sources of supply wherein costs and market prices are subject to time variations and process variables are controllable, each of said plants operating at a different cost function, the method which comprises:
supplying from at least one. of said plants to another of said plants an intermediate product,
generating input signals indicating the input and operating parameters of each of said plants,
applying said input signals to plant-computer-controllers, one plant-computer-controller for each of said plants,
operating said plant-computer-controllers in response to said input signals to automatically produce an operating point signal for each of said plants,
operating each of said plants in response to the operating point signal from the respective plant-computer-controller,
operating a system computer-controller in response to inputs representing the cost and price of materials and finished products to automatically produce system control signals which optimize operation of said system, and
periodically applying said system control signals to said plant-computer-controllers to modify said operating point signals to produce optimized operation of said system as a whole.
The Board
The board reversed a rejection under 35 USC 112, para. 1, finding the disclosure sufficient “within its four corners,” and without aid of appellant’s affidavits, because “one of ordinary skill in the art without undue experimentation would be able to practice the invention” using Deutsch’s “disclosure with its formulas or algorithms * * * together with Phister.’’2
The board affirmed the rejection of the claims as non-statutory under 35 USC 101. It held Gottschalk v. Benson, 409 U.S. 63, 93 S.Ct. 253, 34 L.Ed.2d 273 (1972) to be controlling and cited In re Christensen, 478 F.2d 1392, 178 USPQ 35 (Cust. & Pat.App. 1973), wherein we followed Benson, quoting:
The issue before us in the instant case is also a narrow one, namely, is a method claim in which the point of novelty is a mathematical equation to be solved as the final step of the method, a statutory method? We follow the Supreme Court in concluding that the answer is in the negative.3
Because Deutsch admitted in his brief below that the incorporated Phister reference “shows control and optimization of a multi-unit system employing a particular algorithm or formula of his own,” the board found Deutsch’s “point of novelty” to be his formula or algorithm. Consequently, the *692claims were held non-statutory within the ambit of the cited authorities.
Issue
The sole issue is whether the methods recited in independent claims 1, 6, and 8 constitute non-statutory subject matter under 35 USC 101.4
OPINION
Mathematical problem-solving algorithms were held non-statutory subject matter in Benson.5 Benson was held to have claimed a method for programming a general-purpose digital computer to convert signals from binary-coded decimal form into pure binary form. That method was not “limited to any particular art or technology, to any particular apparatus or machinery, or to any particular end use,” 409 U.S. at 64, 93 S.Ct. at 254, and had “no substantial practical application except in connection with a digital computer,” 409 U.S. at 71, 93 S.Ct. at 257.
In In re Chatfield, 545 F.2d 152, 191 USPQ 730 (Cust. & Pat.App.1976), the majority found the “fundamental rationale” of Benson to be that a method encompassing all practical use of a mathematical formula and the involved algorithm constituted non-statutory subject matter. Chatfield’s method, however, manipulated the machines of a computing machine system for improved efficiency in executing multiple processing programs. Any algorithm used in his method was incidental thereto.
Each of Deutsch’s claimed methods, considered as an industrial process, is even further removed from that of Benson than was that of Chatfield. Each of Deutsch’s methods is a method of operating a system of manufacturing plants. Though Deutsch’s specification teaches that the claimed methods may be carried out by the use of known optimization techniques and apparatus, the claimed system control is exercised externally to a specific optimization or computing technique. The “processing” programs, if such they are, are incidental to the invention.
Unlike the abandoned claims, which included formulae and algorithms, the claims on appeal are drawn to system-operating methods in which system control is applied at less frequent intervals than those at which individual plant controls are applied. Claim 6 states the concept explicitly and claims 1 and 8 do so by calling for periodic application of the system control. Thus, the claimed invention lies in the timing and sequencing of control application, not in the control means (“optimization technique”) itself.
Nothing in the methods claimed by Deutsch preempts a mathematical formula, an algorithm, or any specific computer program. Assuming eventual issue of the claims on appeal, the formulae, algorithms, and programs disclosed in Deutsch’s specification would be freely available to all and could be used for any purpose other than the operation of a system of plants or their equivalent, as spelled out in the appealed claims.
The manufacturing plants upon which Deutsch’s claimed methods operate do use computing “apparatus” to produce finished products from raw materials. Deutsch’s specification does disclose such apparatus as useful in carrying out his methods, and computing “apparatus” includes a “pro*693grammed” digital computer. The specification, however, encompasses alternative means adverted to in appellant’s Beam affidavit:
The “computer-controllers” and “optimizer controller” referred to in [the] specification and drawings are not further characterized as to type except, on p. 15, line 1 where “memory capacity” is referred to; this is usually characteristic of programmable digital computers. Further references to “programming” are on p. 15 of the disclosure, suggesting the inventor’s intent to use a programmable digital computer for the optimization task. It is however fully conceivable that the computer(s) could be “hardwired”, i.e. that programming be implemented via particular circuit configurations. It is also conceivable that an analog computer could be used, in which case the memory referred to (and required, for the specified intermittent operation of the computer) could perhaps also be in analog form. [Emphasis in original.]
Deutsch’s specification included the disclosure of viable alternatives to his preferred optimization technique:
Other optimizing systems and techniques are known in the art and may be employed as alternatives to that shown in Phister [sic] et al. For example, Patent 3,044,701 to Kerstukos et al. describes the method of optimization of a given process unit.
Thus, Deutsch’s specification as a whole reflects, as the board found, the full accomplishment of the specification function, i.e., the enabling of those skilled in the art in the practice of the invention. As above indicated, consideration of the claimed invention as a whole, in the light of the specification, reflects no effort to preempt, and no resulting preemption of, a mathematical formula, an involved algorithm, or a program. Though ably argued by the Solicitor, the questions of whether the programs fed to the computers described in Deutsch’s specification are “processing” programs, and whether programs are themselves patentable, are thus not relevant on this appeal.
A determination of the presence or absence of statutory subject matter must rest upon the claimed invention considered as a whole, not upon selected unclaimed portions of the specification. It was error, therefore, for the board to focus upon one of Deutsch’s disclosed optimization techniques and to transpose it from the specification to the claims as though the optimization technique were being claimed and preempted.
Because the claimed invention considered as a whole does not preempt a mathematical formula, an involved algorithm, or a program per se, and because it is within the technologically useful art of controlling and optimizing a system of manufacturing plants to a particular end use, it is a statutory “process” within the purview of 35 USC 101. In re Foster, 438 F.2d 1011, 58 CCPA 1001 (1971).
The decision of the board is reversed.
REVERSED.
. Phister, Jr., et al. (Phister) U.S. Patent No. 3,079,079 and Kerstukos et al., U.S. Patent No. 3,044,701.
. A paper entitled “SPECIFICATION OF SERIAL NO. 86,296 AS AMENDED” was attached to a supplemental brief submitted after the examiner’s answer but prior to the board’s decision. That amended specification contains no “formulas or algorithms.” Because the status of that paper is unclear from the record and because it was not apparently considered by the board, we do not consider it here.
. In Chatfieid, infra, 545 F.2d at 158, 191 USPQ at 736, decided subsequent to the board’s decision herein, we clarified the passage of Christensen, quoted by the board:
Our reference in Christensen to the mathematical equation as being “at the point of novelty” does not equate to a holding that a claim may be dissected, the claim components searched in the prior art, and, if the only component found novel is outside the statutory classes of invention, the claim amy be rejected under 35 USC 101. That procedure is neither correct nor within the intent of Congress, for the reasons we stated in Bernhart.
. Deutsch acknowledged below that Phister teaches control and optimization of a unit in a multi-unit plant. The claimed methods call for similar control, but for an entire system of multi-unit plants. No rejection under 35 USC 103, however, was entered below. Hence no such rejection may be considered by us at this stage in the prosecution process.
. Methods of doing business and mental steps were discussed at oral argument. Deutsch’s methods are not methods of doing business. See In re Wait, 73 F.2d 982, 22 CCPA 822 (1934). They do not merely facilitate business dealings. That translation of business data into mathematical language intelligible to computers is employed in carrying them out does not make a method of automatically controlling a system of manufacturing plants a method of “doing business.” For that reason, also, the doctrine of mental steps is inapposite. See In re Musgrave, 431 F.2d 882, 57 CCPA 1352 (1970); In re Prater, 415 F.2d 1393, 56 CCPA 1381 (1969).