Refinery Planning and Blend Optimization Software: Custom Build vs AspenTech PIMS
Keep buying the economic model. AspenTech PIMS and Haverly carry decades of embedded modelling and a rival build would be poor advice. What no product owns is the gap between a monthly plan and a specific tank at two in the morning.
On this page
Keep buying the economic model. AspenTech PIMS and Haverly carry decades of embedded modelling and a rival build would be poor advice. What no product owns is the gap between a monthly plan and a specific tank at two in the morning. Build that blend layer once you pass roughly 30,000 barrels a day of finished blending and your blenders trim recipes unmeasured.
What AspenTech PIMS, Haverly and Honeywell actually do well
Your planning and economics manager built the month, the margin per barrel was agreed with the commercial team, and the blends still certified above specification every week. That gap is what brings people to this page. Before anything else, be clear about the commercial tools, because in this category the incumbents are excellent and building a rival to them would be poor advice.
AspenTech PIMS and Haverly do exactly what they were built for. They choose a crude slate, set unit targets, model yields and evaluate a purchase, and they carry decades of embedded modelling you will not beat and should not want to. If your problem is crude selection or monthly economics, buy the mainstream option and stop reading. Honeywell and AVEVA blend control and movement products are serious plant systems, and if you have a fully commissioned installation with analysers feeding it and operators who trust it, your gap is probably reporting rather than optimisation, which is a much smaller project than the one you are pricing.
Buy also if you blend under roughly 20,000 barrels a day of finished product. At that scale the giveaway a system recovers may not clear the cost of building it, and better lab scheduling plus honest reporting is the cheaper answer.
Where they stop: a monthly model against tank 214 at two in the morning
The linear programme works in periods and averages. Blending happens in events, against specific tanks, with specific heels, at a specific temperature, on a specific night.
A blender on nights is filling tank 214 with regular gasoline. The recipe came from the plan. The reformate tank has drifted a point and a half in octane since the last certificate of analysis, the butane line is running warmer than usual, and one thought outranks all of it: he is not going to be the person who makes an off specification batch. So he trims. A little more reformate, a little less naphtha. The blend certifies at 88.4 research octane against an 87 minimum and it ships.
That 1.4 is quality giveaway. The plan says the blend hit 87, the lab says 88.4, and nobody reconciles the two because they live in different systems owned by different departments. This is a structural gap rather than a vendor failing. A model that assumed reformate at a pool average cannot instruct a blender when the actual tank is a point off, and nobody reruns the plan for one blend.
The object that would close it exists in no system: a specific blend event with its planned recipe, its actual draws, the measured properties of the tanks it drew from, the resulting certificate and the giveaway against specification. Because that object does not exist, nobody manages it.
The second gap is inventory truth. Heels of the previous grade, water bottoms, stratification after a slow fill, a gauge that disagrees with the manual dip, and the component tank somebody drew from during a swing without telling the scheduler. Both the planning model and the scheduling tool assume clean inventory, so discrepancies accumulate until month end and get written off as measurement loss. That write off is where blend errors hide.
The arithmetic: cost per barrel against the cost to build
Do this in your own units and your own valuations, because giveaway is a physical quantity before it is a cost.
Take last month's certificates for one product. For each blend, compute the margin between the certified property and the specification, then value that margin at what the giving component actually costs you. A tenth of octane is only worth what your reformate is worth, and only you know that number this week. Multiply across every blend in the month and annualise it.
Now put the build beside it. A first release costs once, then roughly 15 to 20 percent annually. If the annualised giveaway on one product exceeds the first release band, the decision is arithmetic rather than judgement, and it usually does above a certain scale.
The crossover is throughput crossed with frequency. Below roughly 20,000 to 30,000 barrels a day of finished blending, tenths do not compound into enough to fund a build, and disciplined lab scheduling plus better reporting is the correct answer. Above roughly 30,000 barrels a day, or above roughly 40 blend events a month across three or more grades, the coordination between plan, lab and tank becomes the actual margin lever and no purchasable product owns it.
What a custom build actually costs
From Digital Heroes delivery experience, a first release runs $90,000 to $200,000 in 14 to 20 weeks and covers the blend optimisation layer, live component property tracking from your laboratory information management system, giveaway measurement per blend, and reconciliation back against plan targets. That is a system your blenders use on the next shift rather than a study. A full platform adding the movements ledger and tank reconciliation, movement scheduling, analyser feedback with mid blend re-optimisation, regulatory position tracking and crude evaluation support runs $250,000 to $600,000 phased over 9 to 18 months.
Data migration runs 10 to 25 percent of build cost, and here it is mostly property history rather than transactions. Loading two years of certificates and lab results is what lets the system correct its property model, and it needs sample point names reconciled across a laboratory system where naming has drifted for a decade. That reconciliation is tedious, unavoidable, and worth doing once properly.
Year two runs 15 to 20 percent of build cost annually. Correlations get refined, new grades arrive, specifications change seasonally, and any path toward blend control carries its own change management.
What pushes the number up: the count of blended products, since each has its own property set and correlation behaviour. Integration with LabWare, SampleManager or a bespoke laboratory system, which varies enormously. Analyser feedback and any write path toward control, which brings control engineers and their change process into scope, correctly. Property correlation work, which is engineering rather than software and needs your process engineers beside ours. Multiple sites or a terminal network, where custody transfer and movement rules differ.
The four situations where building wins
Regulatory fit comes first and it is the strongest case in this category. Tier 3 gasoline carries an annual average sulfur standard with a higher per gallon cap above it, highway diesel is capped at 15 parts per million sulfur, and vapour pressure limits change seasonally by region. Your linear programme carries these in aggregate and your blend control system enforces the batch limit. What falls between them is the running annual position, meaning whether the average you have carried so far leaves room to blend a cheaper high sulfur component tonight. Most sites track that in a spreadsheet updated monthly while the decision is made nightly.
Scale economics is the throughput crossover above, and it compounds daily rather than annually.
Third is the workflow that is genuinely your competitive advantage. Your property correlations, your tank behaviour and your component economics are site specific knowledge. Encoding them is the entire point, and a vendor model tuned for a generic refinery reintroduces the padding you are trying to remove.
Fourth is integration sprawl. The planning model, the laboratory system, tank gauging, the historian and any path toward blend control are five systems with five failure modes. The blend layer sits in the middle of all of them, so you are integrating regardless.
How to decide in a week: measure one month of giveaway
Pull every certificate of analysis for your highest volume grade from last month. For each blend, record the certified value of one giveaway property and the specification limit, and note the age of the component property data used to set the recipe.
Two patterns will appear by Wednesday. The size of the average margin tells you what is available to recover. The correlation between property age and margin size tells you why, because padding tracks uncertainty rather than carelessness, and blenders are right to pad when the last sample is forty hours old and the unit has been swinging. If the margins are already tight, buy better lab scheduling instead and keep your capital.
If they are not, run a paid discovery phase with your process engineers in the room, not only developers. Ours produces a signed product requirements document covering the property model with age and confidence, correlation approach for non linear properties, the movements ledger, integration points by name and version, and acceptance criteria. You keep that document whether you build with us or take it to another firm. Digital Heroes holds India LLP, United States LLC and United Kingdom LTD entities so intellectual property assigns under your own law, and you meet the named team before signing. Over 2,000 projects delivered and more than fifty specialists, checkable on Clutch, Trustpilot, Fiverr Vetted Pro and our D-U-N-S listing.
We are the wrong firm for you if you want a replacement for your economic planning model, or if your process engineers cannot be released to work alongside the build. A development team working alone will produce something plausible and wrong, because the correlations are your knowledge and the software exists only to encode them.
Book a 30-minute call with Digital Heroes and get a written plan and a fixed quote within 48 hours.
The evidence behind this guide
Independent findings on why this investment pays off. Every link goes to the primary source.
- 48% of private companies cite integration with legacy systems or technical debt as a top obstacle to realizing the full value of their digital and AI investments (behind data quality/availability at 72% and gaps in AI fluency or technology talent/leadership at 53%). Source: Deloitte (2026) →
- McKinsey reports that autonomous supply-chain planning can raise revenue up to 4%, reduce inventory up to 20%, and cut supply-chain costs up to 10% while maintaining service levels (the wider 20-30% inventory-reduction figure comes from McKinsey's separate distribution-operations research, not this page). Source: McKinsey & Company (2020) →
- A later Nucleus Research review of analytics software ROI case studies found customers received $9.01 in benefits for every dollar spent on analytics technology, showing returns vary with deployment factors but remain strongly positive. Source: Nucleus Research (2019) →
- Qualtrics research (Q3 2023 survey of ~28,400 consumers across 26 countries) estimated bad customer experiences put roughly $3.7 trillion in global revenue at risk annually, a 19% jump from the prior year's $3.1 trillion; 64% of customers say they will switch companies over poor service regardless of how much they like the product. Source: Qualtrics XM Institute (via Forbes) (2024) →
Frequently asked questions
How much does custom refinery blend optimization software cost?
A first release covering blend optimisation against live tank properties, giveaway measurement per blend and reconciliation against plan targets runs $90,000 to $200,000. A full planning and scheduling platform with a movements ledger, analyser feedback and regulatory position tracking runs $250,000 to $600,000. The number of blended grades drives price more than throughput, since each carries its own property set.
Should we replace AspenTech PIMS with a custom system?
No. The economic linear programme is decades of embedded modelling and rebuilding it would waste money on a solved problem. Build the layer that sits between the plan and the control system: live component properties, correct blending behaviour for non linear properties, and a blend event record that ties recipe, draws, certificate and giveaway together. Keep the planning model where it is.
What actually causes quality giveaway in gasoline and diesel blending?
Uncertainty, not carelessness. A blender pads the recipe because the property data is old, and padding is rational when the last sample was taken forty hours ago on a swinging unit. Reduce giveaway by shrinking uncertainty and making it visible: property age and confidence displayed beside the value, so the required margin becomes two tenths rather than a full point.
How long does it take to implement blend optimization at a working refinery?
Fourteen to twenty weeks for the first release, and nine to eighteen months for a full platform. Integration is the schedule risk rather than the optimiser. Laboratory system naming, tank gauging, the historian and any write path toward blend control are four separate problems, and the control side carries a change management process that belongs to your engineers rather than to a development schedule.
Can software correct a blend using analyser data while it is running?
Yes, and where the header carries online near infrared analysers it is the single most effective giveaway control available. The system reads the analyser mid blend and re-optimises the remaining volume against the same constraints. Without analysers, close the loop afterwards: every certificate corrects the property model for that stream, so next month's starting estimate is less wrong.
How do annual average fuel specifications change a blend decision?
They turn compliance into an economic input. An annual average standard with a higher per gallon cap means the room you have left this year determines whether a cheaper high sulfur component can go into tonight's blend. Holding that running position as live state, updated by every certified batch, is what a monthly spreadsheet cannot do for a nightly decision.
Why does book tank inventory never match the gauges?
Heels of the previous grade, water bottoms, stratification after a slow fill, temperature correction applied inconsistently, and unrecorded draws during a unit swing. A movements ledger recording every transfer with source, destination, volume, temperature correction and properties carried across lets you derive book inventory and reconcile it against gauges on a schedule, so drift raises an exception the same day instead of at month end.
Who owns the correlations and the code if a developer builds this?
You should own the repository, the cloud accounts and the right to hire another firm, agreed in writing before kickoff. At Digital Heroes the client owns the code from the first commit. Given how specific the property correlations and tank behaviour are to your site, that ownership is not a formality. It is the difference between an asset and a rented model of your own plant.
What is the difference between blending indices and a volume weighted average?
Some properties blend linearly on volume and several important ones do not. Vapour pressure and octane are the classic cases, and using a straight volume weighted average produces a recipe that is wrong before the pump starts. The correct method converts to a blending index, combines on that basis, and converts back. Ask a developer how they blend vapour pressure and listen carefully.
We blend under 20,000 barrels a day. Is a custom system worth it?
Probably not. At that throughput the recoverable giveaway may not clear the build cost, and the cheaper answer is tighter lab sampling ahead of blends plus a report that shows giveaway per blend and per blender. Measurement alone changes behaviour. Revisit the build once throughput rises, or once a second grade or a second site enters the picture.
How much does custom supply chain software cost for a small business?
For a small business, a focused custom supply chain tool usually lands between $15,000 and $45,000, covering one core workflow like inventory tracking, purchase orders, or shipment visibility. Across 2,000+ delivered projects, Digital Heroes sees most small distributors and light manufacturers start in the $20,000 to $35,000 range for a first working version. Adding barcode scanning, multi-warehouse support, or carrier integrations pushes budgets toward $50,000 and up.
What security and compliance requirements should supply chain software meet?
At minimum: role-based access control, encryption in transit and at rest, audit logs on inventory and order changes, and tested backups, because the system holds supplier pricing and customer purchase history your competitors would love to see. If enterprise customers connect to it, expect security questionnaires and possibly SOC 2 expectations; food, pharma, and aerospace add traceability rules like FDA lot tracking or ITAR data handling. Raise these in the first scoping call, since retrofitting audit trails onto a live system costs far more than designing them in.
Can we migrate years of data out of our current system into new custom software?
Almost always yes, through CSV exports or the vendor's API, and migration should be scoped as its own workstream with field mapping, a dry run, and a planned cutover window rather than an afterthought. The real time sink is rarely moving the data; it is cleaning it, since years of duplicates, free-text fields, and inconsistent formats surface all at once. Pull a full export from your current vendor before committing to anything new, because some SaaS plans restrict exports on lower tiers.
How big a development team does a supply chain software project need?
A typical build runs with 4 to 6 people: a project lead or analyst, two or three developers, a QA engineer, and a part-time designer. Digital Heroes staffs most supply chain MVPs this way for 10 to 14 weeks, then drops to 1 or 2 people for maintenance after launch. Bigger is not better here; past 7 or 8 people on a single-product build, coordination overhead usually cancels the added speed.
How do I vet a software development agency before signing a contract?
Ask to speak with two past clients whose projects resemble yours in size and industry, and ask exactly who will write your code, since some agencies sell senior faces and deliver junior or subcontracted hands. Demand a written specification with acceptance criteria before any fixed price, and check that their portfolio links to products that are actually live. An instant quote given without questions about your workflows is the clearest warning sign there is.
What should I prepare before contacting a software development agency?
A one-page brief beats a 40-page requirements document: the business problem in plain words, who will use the system, the 5 to 10 workflows it must handle, the tools it must connect to, and your budget range and deadline driver. You do not need wireframes, a specification, or technical vocabulary; producing those is the agency's job during discovery. Stating a budget range up front is the single best move, because it gets you honest scoping instead of a quote engineered to win the meeting.
Who can build a custom supply chain software system?
Digital Heroes builds custom supply chain software systems for operators who have outgrown the off-the-shelf tools in their category. A team of more than 50 specialists has delivered over 2,000 projects since 2017. Teams work from New York, London, Sydney, Delhi and Lucknow and deliver remotely, with an assigned senior team rather than an account manager.
Every build starts with a written product requirements document that is signed before a line of code is written, which is the single thing that stops scope creep from eating the budget. Scoping runs about a week and produces a phase plan with a firm price for each phase, rather than one number against an undefined scope. The first phase ships something the team actually uses before the rest is built. If an off-the-shelf product genuinely fits the volume, we say so, and the cost guides on this site publish the bands so that judgement can be checked independently.
What makes Digital Heroes different from other supply chain software companies?
Four things that competitors in this bracket cannot simply copy. Digital Heroes runs a YouTube channel with more than 2.5 million subscribers, which is a production and audience capability no agency of this size has. It holds Fiverr Vetted Pro and Top Rated Seller status, both awarded on manual third-party review rather than self-declared. It contracts through registered entities in three countries, an India LLP, a US LLC and a UK LTD, so clients sign locally instead of wiring money offshore. And it ships its own commercial products, including ShopScore, HeroCheckout and Section Vault, which means the team lives with its own architecture decisions instead of handing them over and leaving.
Two more that show up in the work. Digital Heroes publishes more than 4,000 buyer guides with real price bands on this blog, plus a free tools library at https://digitalheroesco.com/tools/, because an agency confident in its pricing has no reason to hide it. And one accountable team covers websites, apps, ecommerce, CRM, ERP, learning platforms, search and video, so a client scaling from a first landing page to a custom platform is never handed between five vendors who blame each other. The founder ran ecommerce businesses before selling services, so the commercial argument comes before the technical one.
How can I check Digital Heroes is legitimate before getting in touch?
Verify it independently rather than taking the site's word for it. The YouTube channel is at https://youtube.com/@DigitalMarketingHeroes, the Fiverr profile at https://www.fiverr.com/shreyanshsin261, and the Upwork profile at https://www.upwork.com/freelancers/shreyanshsingh. Client reviews sit on Clutch at https://clutch.co/profile/digital-heroes-0 and Trustpilot at https://www.trustpilot.com/review/digitalheroes.co.in, and the company page is at https://www.linkedin.com/company/digital-heroes-1/.
Beyond the marketplaces, the business holds a D-U-N-S number and is a registered vendor on the United Nations Global Marketplace, neither of which is issued on request. Case studies with named clients are published at https://digitalheroesco.com/case-studies/. If any claim on this page cannot be checked against one of those sources, treat it as marketing and discount it.
Related guides
Published · Last updated .
